DESC:
FORM 2
THE PATENTS ACT, 1970
(39 OF 1970)
&
THE PATENTS RULES, 2003

COMPLETE SPECIFICATION
[SEE SECTION 10, RULE 13]

A SYSTEM AND METHOD FOR SHUTTERLESS TEMPERATURE-ROBUST CALIBRATION OF INFRARED SENSORS

BHARAT ELECTRONICS LIMITED
WITH ADDRESS: OUTER RING ROAD, NAGAVARA, BANGALORE 560045, KARNATAKA, INDIA

THE FOLLOWING SPECIFICATION PARTICULARLY DESCRIBES THE INVENTION AND THE MANNER IN WHICH IT IS TO BE PERFORMED.

TECHNICAL FIELD
[0001] The present invention relates generally to a field of infrared imaging systems. In particular, the present invention relates to systems and methods for calibrating infrared (IR) sensors used in thermal imaging devices.
BACKGROUND
[0002] With the advancements in an infrared sensor technology, thermal imaging devices have become a widely used tool for imaging in dark and diverse environments. Typical applications of thermal imagers include autonomous systems like an advanced driver-assistance system (ADAS), defense applications, industrial applications, and medical applications as well.
[0003] As a fabrication technology for IR sensors is not as mature as that of CCD (charge-coupled device) or CMOS (complementary metal-oxide semiconductor) sensors, these IR sensors often suffer from various types of noise, including spatial and temporal noise. The spatial noise is characterized by variation in a response of an individual IR sensor spread across a focal plane array, when exposed to a uniform thermal radiation. This type of spatial noise is also referred to as fixed pattern noise in the art because such noise generally does not change (under certain conditions). Since the noise remains fixed, it is possible to suppress it with some calibration technique. In a context of IR sensors such fixed pattern noise is also known as non-uniformity in the art, and hence the calibration technique to suppress this noise is also known as non-uniformity correction (NUC) or flat field correction. The fixed pattern noise does not change as long as the temperature of the IR sensor is maintained fixed. The class of infrared imaging devices which keeps the IR sensor at fixed known temperature are called as cooled IR imaging devices, whereas the ones in which the IR sensor temperature varies with ambient temperature are called as un-cooled IR imaging devices. The cooled IR devices maintain the temperature of the IR sensor at around 77K with a cryogenic cooler attached to the IR sensor. However, because of design and manufacturing imperfections, there are certain temperature fluctuations in the IR sensor of the cooled devices as well. The uncooled devices on the other hand does not have mechanism to control IR sensor temperature, and the sensor temperature varies with the ambient temperature. Under such scenarios, the fixed pattern noise no longer remains fixed and keeps on changing as the temperature of IR sensor changes.
[0004] In the art, the techniques for suppressing spatial non-uniformity in IR sensors can be divided into two broad classes, viz. calibration based, and scene based. The calibration based techniques compute some calibration parameters by exposing the IR sensor to a known environment under controlled settings, and then use these calibration parameters for correcting the non-uniformity. Whereas, in scene based techniques, the non-uniformity is corrected by estimating some scene statistics on the go and using these scene statistics to correct sensor non-uniformity. The calibration based techniques can be further divided into two sub-classes viz. laboratory calibration and shutter calibration. In the laboratory calibration techniques, the IR sensor is exposed to a uniform radiating surface maintained at different temperatures. In the shutter calibration techniques, the IR sensor is exposed to a uniform scene (i.e., a shutter) at some unknown temperature. To mitigate the variation in fixed pattern noise due to variation IR sensor temperature, the calibration process is performed in a controlled environment, maintaining the IR sensor temperature at various known values. The calibration parameters for various IR sensor temperature are stored and used later for correcting the non-uniformity.
[0005] In calibration based techniques, a mechanism to obtain temperature of the IR sensor is required. Such a mechanism is not always available. Furthermore, storing hundreds of calibration tables (for various IR sensor temperature) onboard the IR sensor requires huge amount of an onboard memory. For a shutter based calibration technique, a mechanical shutter needs to be installed with the IR sensor. These additional constraints / requirements increase the cost, and design complexity of the IR imaging devices. One such technique is disclosed in the art US9491376. Such techniques performs well but at the cost of increased design complexity and expenditure. On the other hand, scene based calibration techniques do not put specific design constraint on the imaging device, but they do put constraint on the operating environment. Specifically, scene based calibration techniques generally require some kind of blur in the scene which is either induced by motion in the scene or by movement of lens or the complete imaging device. One such technique is disclosed in the art WO2012170949A2.
[0006] Hence, there is a need of improvement in the calibration techniques for IR sensor, to make them more robust to variation in IR sensor temperature, while keeping the imaging system and operating constraints less complex and cost effective.
SUMMARY
[0007] This summary is provided to introduce concepts related toa system and method for shutterless temperature-robust calibration of infrared sensors. This summary is neither intended to identify essential features of the present invention nor is it intended for use in determining or limiting the scope of the present invention.
[0008] For example, various embodiments herein may include one or more systems and methods thereof are provided. In one of the embodiments, a method for shutterless temperature-robust calibration includes a step of sensing, by at least one sensor, electromagnetic radiations in a wavelength band and generating an image based on the sensed radiations. The method includes a step of setting, by a black body radiator, a pre-determined temperature for a black body radiator surface. The method includes a step of acquiring, by an acquisition module, a plurality of image data samples at a pre-defined time interval from the generated image. The method includes a step of processing, by a processing module, the acquired image data samples and generating at least one calibration parameter. The method includes a step of estimating, by a noise parameter estimation module of the processing module, noise parameters from the acquired image data samples. The method includes a step of correcting, by a correction module, the acquired data samples based on the estimated noise parameters and generating corrected image data samples.
[0009] In another embodiment, a system for shutterless temperature-robust calibration includes at least one sensor, a black body radiator, an acquisition module, a processing module, and a correction module. The at least one sensor is enclosed in a thermal chamber with a window. The sensor is configured to sense electromagnetic radiations in a wavelength band and generate an image based on the sensed radiations. The black body radiator is placed facing the window of the thermal chamber and is further configured to set a pre-determined temperature for a black body radiator surface. The acquisition module is configured to acquire a plurality of image data samples at a pre-defined time interval from the generated image. The processing module is configured to process the acquired image data samples and generate at least one calibration parameter. A noise parameter estimation module of the processing module is further configured to estimate noise parameters from the acquired image data samples. The correction module is configured to correct the acquired data samples based on the estimated noise parameters and generate corrected image data samples.
BRIEF DESCRIPTION OF ACCOMPANYING DRAWINGS
[0010] The detailed description is described with reference to the accompanying figures. In the figures, the left-most digit(s) of a reference number identifies the figure in which the reference number first appears. The same numbers are used throughout the drawings to reference like features and modules.
[0011] Figure 1 illustrates a block diagram depicting a system for shutterless temperature-robust calibration, according to an implementation of the present invention.
[0012] Figure 2A illustrates a block diagram depicting a system for shutterless temperature-robust calibration of infrared sensors, according to an exemplary implementation of the present invention.
[0013] Figure 2B illustrates a schematic diagram depicting a system for shutterless temperature-robust calibration of infrared sensors for correction, according to an exemplary implementation of the present invention.
[0014] Figure 3 illustrates a schematic diagram depicting control and data acquisition, according to an exemplary implementation of the present invention.
[0015] Figure 4 illustrates a schematic diagram depicting a processing module of Figure 1, according to an exemplary implementation of the present invention.
[0016] Figure 5 illustrates a schematic diagram depicting noise parameter estimation, according to an exemplary implementation of the present invention.
[0017] Figure 6 illustrates a flow diagram depicting a method for calibrating infrared sensors, according to an exemplary implementation of the present invention.
[0018] Figure 7 illustrates a flowchart depicting a method for shutterless temperature-robust calibration of infrared sensors, according to an implementation of the present invention.
[0019] It should be appreciated by those skilled in the art that any block diagrams herein represent conceptual views of illustrative systems/platforms embodying the principles of the present invention. Similarly, it will be appreciated that any flowcharts, flow diagrams, and the like represent various processes which may be substantially represented in computer readable medium and so executed by a computer or processor, whether or not such computer or processor is explicitly shown.
DETAILED DESCRIPTION
[0020] In the following description, for the purpose of explanation, specific details are set forth in order to provide an understanding of the present invention. It will be apparent, however, to one skilled in the art that the present invention may be practiced without these details. One skilled in the art will recognize that embodiments of the present invention, some of which are described below, may be incorporated into a number of systems.
[0021] The various embodiments of the present invention provide a system and method for shutterless temperature-robust calibration of infrared sensors. Furthermore, connections between components and/or modules within the figures are not intended to be limited to direct connections. Rather, these components and modules may be modified, re-formatted or otherwise changed by intermediary components and modules.
[0022] References in the present invention to “one embodiment” or “an embodiment” mean that a particular feature, structure, characteristic, or function described in connection with the embodiment is included in at least one embodiment of the invention. The appearances of the phrase “in one embodiment” in various places in the specification are not necessarily all referring to the same embodiment.
[0023] In one of the embodiments, a method for shutterless temperature-robust calibration includes a step of sensing, by at least one sensor, electromagnetic radiations in a wavelength band and generating an image based on the sensed radiations. The method includes a step of setting, by a black body radiator, a pre-determined temperature for a black body radiator surface. The method includes a step of acquiring, by an acquisition module, a plurality of image data samples at a pre-defined time interval from the generated image. The method includes a step of processing, by a processing module, the acquired image data samples and generating at least one calibration parameter. The method includes a step of estimating, by a noise parameter estimation module of the processing module, noise parameters from the acquired image data samples. The method includes a step of correcting, by a correction module, the acquired data samples based on the estimated noise parameters and generating corrected image data samples.
[0024] In another implementation, the method includes a step of storing, by two first memories, a plurality of temperature values for configuring the thermal chamber and the black body radiator, respectively. The method includes a step of storing, by a second memory, the acquired image data samples. The method includes a step of generating, by a logic module, an output logic state based on the acknowledgement signals. The method includes a step of delaying, by a delay module, the output logic state for stabilizing the temperature of the sensor to a configured temperature value.
[0025] In another implementation, the method includes a step of reading, by the acquisition module, the temperature values from the first memory and generating and transmitting a first configuration command to the thermal chamber for configuration based on the read temperature values.
[0026] In another implementation, the method includes a step of reading, by the acquisition module, the temperature values from the first memory and generating and transmitting a second configuration command to the black body radiator for configuration based on the read temperature values.
[0027] In another implementation, the step of processing the acquired data samples and generating the calibration parameter further includes a step of counting, by a counter, a number of image data samples of the generated image; reading, by a step integrator, at least one image data sample and generating an integrated image data sample; reading, by a divider, the integrated image data sample and generating an output data sample; performing, by a regression module, a regression analysis on the image data samples and the output data sample and generating the calibration parameter; and estimating, by the noise parameter estimation module, noise parameters from the acquired image data samples.
[0028] In another implementation, the step of estimating the noise parameter further includes a step of receiving, by a subtractor module, the acquired image data samples; generating, by the subtractor module, an output by using the integrated image sample based on the received acquired image data samples; squaring, by a squaring module, the generated output and generating a squared output; accumulating, by an accumulator module, the squared output and the new incoming image data samples and generating an accumulated output; computing, by a running mean computing module, a median based on the accumulated output; and estimating, by the running mean computing module, the noise parameters based on the computed median.
[0029] In another embodiment, a system for shutterless temperature-robust calibration includes at least one sensor, a black body radiator, an acquisition module, a processing module, and a correction module. The at least one sensor is enclosed in a thermal chamber with a window. The sensor is configured to sense electromagnetic radiations in a wavelength band and generate an image based on the sensed radiations. The black body radiator is placed facing the window of the thermal chamber and is further configured to set a pre-determined temperature for a black body radiator surface. The acquisition module is configured to acquire a plurality of image data samples at a pre-defined time interval from the generated image. The processing module is configured to process the acquired image data samples and generate at least one calibration parameter. A noise parameter estimation module of the processing module is configured to estimate noise parameters from the acquired image data samples. The correction module is configured to correct the acquired data samples based on the estimated noise parameters and generate corrected image data samples.
[0030] In another implementation, the sensor is enclosed in the thermal chamber in a manner such that the sensor faces the window towards the black body radiator and a field-of-view of the sensor falls within a perimeter of the black body radiator surface.
[0031] In another implementation, the sensor is an infrared (IR) sensor.
[0032] In another implementation, the acquisition module includes two first memories, a second memory, a logic module, and a delay module. The two first memories are configured to store a plurality of temperature values for configuring the thermal chamber and the black body radiator, respectively; The second memory is configured to store the acquired image data samples. The logic module is configured to receive acknowledgement signals from the thermal chamber and black body radiator, which is further configured to generate an output logic state based on the acknowledgement signals. The delay module is configured to delay the output logic state to stabilize the temperature of the sensor to a configured temperature value. The first memories are non-volatile memories, and the second memory is a volatile memory.
[0033] In another implementation, the acquisition module is configured to read the temperature values from the first memory and generate and transmit a first configuration command to the thermal chamber for configuration based on the read temperature values.
[0034] In another implementation, the acquisition module is configured to read the temperature values from the first memory and generate and transmit a second configuration command to the black body radiator for configuration based on the read temperature values.
[0035] In another implementation, the processing module includes a counter, a step integrator, a divider, and a regression module. The counter is configured to count a number of image data samples of the generated image. The step integrator is configured to read at least one image data sample and generate an integrated image data sample. The divider is configured to read the integrated image data sample and generate an output data sample. The regression module is configured to perform a regression analysis on the image data samples and the output data sample and generate the calibration parameter. The noise parameter estimation module is configured to estimate noise parameters from the acquired image data samples.
[0036] In another implementation, the noise parameter estimation module includes, a subtractor module, a squaring module, an accumulator module, and a running mean computing module. The subtractor module is configured to receive the acquired image data samples, and is further configured to generate an output by using the integrated image sample. The squaring module is configured to square the generated output and generate squared output. The accumulator module is configured to accumulate the squared output and the new incoming image data samples and generate an accumulated output. The running mean computing module is configured to compute a median based on the accumulated output and estimate the noise parameters based on the computed median.
[0037] In an exemplary embodiment, the present invention discloses a novel calibration and correction technique for infrared (IR) sensors, which is robust to variation in IR sensor temperature, simple and efficient in terms of realization in an IR sensor, and cost effective.
[0038] In another embodiment, the present invention discloses a system and method for calibration and correction of temperature varying spatial non-uniformity in IR sensors used in thermal imaging devices. Unlike prior arts, the present invention neither requires IR sensor temperature and multiple calibration tables for various IR sensor temperatures, nor does it require any mechanical shutter or any constraints on the scene being imaged. The present invention proposes to estimate a noise parameter from image statistics. This noise parameter is a representative of IR sensor temperature and resulting non-uniformity. The present invention models this noise parameter and computes the parameters of this model (i.e., calibration parameters) during calibration. For this purpose, the present invention discloses a method for calibration using a thermal chamber and black body radiator. For correcting the IR sensor captured data, the present invention computes noise parameter and uses model parameters (i.e., calibration parameters) obtained during calibration to generate corrected image data.
[0039] In an embodiment, the present invention discloses a system and method for calibration and correction of IR sensors data samples to suppress temperature varying spatial non-uniformity. Prior arts for suppressing / correcting temperature varying spatial non-uniformity in IR sensors, involves some calibration based techniques which perform multiple calibrations with a shutter or with a black body at multiple IR sensor temperatures, and some scene based techniques which require blur in the scene being imaged. The requirement of shutter or calibrations at multiple IR sensor temperature makes the system design complex and less cost effective, while having constrains on the scene limits the practicality of system. The method and system presented in the present invention neither requires IR sensor temperature or multiple calibration tables for multiple IR sensor temperatures, nor does it require any mechanical shutter or any constraints on the scene being imaged. The present invention estimates a noise parameter from the image statistics. This noise parameter estimate is a representative of IR sensor temperature and resulting spatial noise. The present invention proposes a method for modelling the variation of this noise parameter and uses this model for correcting spatial non-uniformity in IR sensor. Further, unlike many prior arts, the present invention does not require any mechanical shutter or motion / blur constraint on the scene being imaged.
[0040] In an embodiment, the present invention relates generally to the field of infrared imaging systems. In particular, the present invention relates to systems and methods for calibrating an IR sensor used in thermal imaging devices. The present invention presents system and method for calibrating IR sensors in way which is more efficient in terms of realization in an IR sensor and is not affected by change in temperature of the IR sensor. Prior works depend on the idea of calibrating the IR sensor at different values of IR sensor temperatures, storing all these calibration parameters obtained at different IR sensor temperatures on-board the IR sensor, and then depending on temperature of IR sensor at any specific instant of time using a corresponding set of calibration data at that time for correcting the image data samples captured by IR sensor. Unlike prior works, the invention disclosed here does not need to compute and store calibration data for different IR sensor temperature values. Instead, the present invention disclosed uses a system and method which calibrates the IR sensor and computes calibration parameters in a way such that the computed calibration data becomes suitable for correction at any IR sensor temperature values. The system and method require to store only a few calibration parameters on-board IR sensor, and uses the same calibration parameters at any IR sensor temperature for correcting the image data samples captured by IR sensor. Further, the system and method does not require temperature of IR sensor. Hence, the system and method offer a simple and more efficient way for calibrating and correcting IR sensors which is robust to changes in IR sensor temperature.
[0041] Figure 1 illustrates a block diagram depicting a system (100) for shutterless temperature-robust calibration, according to an implementation of the present invention.
[0042] A system for shutterless temperature-roust calibration (hereinafter referred to as “system) (100) includes at least one sensor (102), a thermal chamber (104), a black radiator (106), an acquisition module (108), a processing module (118), and a correction module (138).
[0043] The at least one sensor (102) is enclosed in the thermal chamber (104) with a window. The sensor (102) is configured to sense electromagnetic radiations in a wavelength band received from a plurality of sources and generate an image based on the sensed radiations. In an embodiment, the sensor (102) is enclosed in the thermal chamber (104) in a way that the sensor (102) looks out the window towards the black body radiator (106) and a field-of-view of the sensor (102) falls within a perimeter of the black body radiator surface. In one embodiment, the sensor (102) is an infrared (IR) sensor. In an exemplary embodiment, the thermal chamber (104) refers to any electronic/electrical/electromechanical device consisting of hardware and/or software with an enclosed thermally insulated volume capable of receiving configuration commands from the external world and maintaining a configured temperature in the enclosed volume consistently for a period of time. In another exemplary embodiment, the sensor (102) refers to any digital device consisting of hardware and/or software capable of sensing electromagnetic radiations in infrared wavelength band and generating an image of a real world scene as observed in that wavelength band. This can also be referred to as infrared focal plane array in the art. In another exemplary embodiment, image data samples refer to digital data or digital representation of analog electrical signals coming via analog-to-digital converter from each of the detector or sensor (102) wherein output of individual detectors are multiplexed to the Read-Out Integrated Circuit and represent the intensity measured by the detector and vary from image source to image source having a fixed range depending on the sensor (102).
[0044] The black body radiator (106) placed at facing the window of the thermal chamber (104). The black body radiator (106) is further configured to set a pre-determined temperature for a black body radiator surface. In an embodiment, the blackbody radiator (106) is configured to emit a maximum amount of thermal radiation possible for an object at a given temperature. In an exemplary embodiment, the black body radiator is any electronic/electrical/electromechanical device consisting of hardware and/or software with a flat black body radiator surface capable of receiving a configuration command from the external world and maintaining a configured temperature on the black body radiator surface consistently throughout the surface for a period of time.
[0045] The acquisition module (108) is configured to cooperate with the sensor (102), the thermal chamber (104), and the black body radiator (106). The acquisition module (108) is configured to acquire a plurality of image data samples at a pre-defined time interval from the generated image.
[0046] In an embodiment, the acquisition module (108) includes two first memories (110), a second memory (112), a logic module (114), and a delay module (116).
[0047] The two first memories (110) are configured to store a plurality of temperature values for configuring the thermal chamber (104) and the black body radiator (106), respectively. In an embodiment, the acquisition module (108) is configured to read the temperature values from the first memory (110) and generate and transmit a first configuration command to the thermal chamber (104) for configuration based on the read temperature values. In another embodiment, the acquisition module (108) is configured to read the temperature values from the first memory (110) and generate and transmit a second configuration command to the black body radiator (106) for configuration based on the read temperature values. In yet another embodiment, the first memories (110) are non-volatile memories. The second memory (112) is configured to store the acquired image data samples. In an embodiment, the second memory (112) is a volatile memory. The logic module (114) is configured to receive acknowledgement signals from the thermal chamber (104) and the black body radiator (106), the logic module configured to generate an output logic state based on the acknowledgement signals. The delay module (116) is configured to delay the output logic state to stabilize the temperature of the sensor (102) to a configured temperature value.
[0048] The processing module (118) is configured to cooperate with the acquisition module (108) to receive the acquired image data samples. The processing module (118) is further configured to process the acquired image data samples and generate at least one calibration parameter and estimate noise parameters.
[0049] In an embodiment, the processing module (118) includes a counter (120), a step integrator (122), a divider (124), a regression module (126), and a noise parameter estimation module (128). The counter (120) is configured to count a number of image data samples of the generated image. The step integrator (122) is configured to read at least one image data sample and generate an integrated image data sample. The divider (116) is configured to read the integrated image data sample and generate an output data sample. The regression module (126) is configured to perform a regression analysis on the image data samples and the output data sample and generate the calibration parameter. The noise parameter estimation module (128) is configured to receive the acquired image data. The noise parameter estimation module (128) is further configured to estimate noise parameters from the acquired image data samples. This noise parameter is then used for generating calibration parameter. In one embodiment, the term “calibration” refers to any method for reading image data samples captured by the IR sensor (102) under laboratory conditions and using them to generate calibration parameters which can be used later for correcting image data samples captured by the IR sensor (102). The term “calibration” can also be referred to as Non-Uniformity Correction (NUC) calibration in the art.
[0050] In an embodiment, the noise parameter estimation module (128) includes a subtractor module (130), a squaring module (132), an accumulator module (134), and a running mean computing module (136).
[0051] The subtractor module (130) is configured to receive the acquired image data samples. The subtractor module (130) is further configured to generate an output by using the integrated image sample. The squaring module (132) is configured to cooperate with the subtractor module (130) to receive the generated output. The squaring module (132) is further configured to square the generated output and generate a squared output. The accumulator module (134) is configured to cooperate with the squaring module (132) to receive the generated squared output. The accumulator module (134) is further configured to accumulate the squared output and the new incoming image data samples and generate an accumulated output. The running mean computing module (136) is configured to cooperate with the accumulator module (134) to receive the accumulated output. The running mean computing module (136) is further configured to compute a median based on the accumulated output and estimate the noise parameters based on the computed median.
[0052] The correction module (138) is configured to cooperate with the noise parameter estimation module (128) to receive the estimated noise parameters. The correction module (138) is further configured to correct the acquired data samples based on the estimated noise parameters and generate corrected image data samples. In an exemplary embodiment, the term "correction" refers to any method for reading a pre-computed/ generated calibration parameters and using them to minimize spatial noise / perturbation / deviations in the image data samples captured by the sensor (102). This can also be referred to as non-uniformity correction or flat field correction in the art.
[0053] In an exemplary embodiment, the system (100) first configures the thermal chamber (104) with the IR sensor (102) placed inside the thermal chamber (104) and the black body radiator with the IR sensor (102) facing towards radiating surface of the black body radiator (106) to set some specific temperatures and captures multiple image data samples at multiple instants of time from the IR sensor (102). This is repeated for different combinations of the thermal chamber (104) and the black body radiator (106) temperature values. Thereafter, the captured image data samples are processed in the step integrator (122) of the processing module (118) to generate the step integrated image data samples. From these step integrated image data samples, a noise parameter is estimated by using the noise parameter estimation module (128). This noise parameter is an indicator of amount of noise present in the IR sensor image data samples which needs to be corrected. Further, by using the image data samples and noise parameter estimation, the regression analysis (126) is carried out to generate calibration parameters. These calibration parameters are then stored on-board IR sensor for correcting the image data samples captured by the IR sensor (102).
[0054] Figure 2A illustrates a block diagram (200A) depicting a system for shutterless temperature-robust calibration of infrared sensors of Figure 1, according to an exemplary implementation of the present invention.
[0055] In Figure 2A, the sensor (along with optics) (102) to be calibrated is kept inside the thermal chamber (104) with a window. The sensor (102) along with optics is placed such that it is field-of-view falls within the perimeter of the window. Opposite to the thermal chamber (104), the black body radiator (106) is placed such that the radiating surface of the black body radiator (106) faces the window of the thermal chamber (104). With this arrangement, the field-of-view of the sensor (102), oriented towards the black body radiator (106) falls within the perimeter of radiating surface of the black body radiator (106). A control and data acquisition electronics (206) consists of electronics for controlling the black body radiator (106), the thermal chamber (104), and the sensor (102). In an embodiment, the control and data acquisition electronics (206) is the acquisition module (108) as shown in the Figure 1. The control and data acquisition electronics (206) also consists of electronics for capturing multiple image data samples at multiple instants of time from the sensor (102). The control and data acquisition electronics (206) configures the black body radiator (106) and the thermal chamber (104), then captures multiple image data samples at multiple instants of time from the sensor (102). The control and data acquisition electronics (206) repeats the sequence of configuring and data capturing for multiple times and writes captured data to a non-volatile memory (202). In an embodiment, the non-volatile memory (202) is the first memory (110) as shown in the Figure 1. The control and data acquisition electronics (206) then triggers the processing electronics (204). The processing electronics (204) consists of electronics for processing the image data samples captured by the control and data acquisition electronics (206). In an embodiment, the processing electronics (204) is the processing module (118) as shown in the Figure 1. The processing electronics (204) reads image data samples from the non-volatile memory (202), processes these samples, and writes the processed data samples to the non-volatile memory (202). The processed data samples are the output of the system (100).
[0056] Figure 2B illustrates a schematic diagram (200B) depicting a system for shutterless temperature-robust calibration of infrared sensors for correction, according to an exemplary implementation of the present invention.
[0057] In Figure 2B, the processed data samples are then used in real-time in the sensor (102) for correcting the real-time captured image data samples, (208). The pre-computed/ generated calibration parameters are stored in the non-volatile memory (220). The image data samples captured by the sensor (102) are acquired with a sensor read-out circuitry (212). These image data samples are written onto an onboard volatile memory (214). In an embodiment, the volatile memory (214) is a second memory (112) as shown in the Figure 1. An on-board noise parameter estimation module (128) computes an estimate of noise at a block (216) present in the captured image data samples. Thereafter, a correction module (138) of the Figure 1 corrects the image data samples, at a block (218), by using the estimated noise parameter and calibration parameters to correct image data samples and generate corrected image data samples.
[0058] Figure 3 illustrates a schematic diagram (300) depicting control and data acquisition, according to an exemplary implementation of the present invention.
[0059] In an embodiment, the acquisition module (108) of the system (100) as shown in the Figure 1 is configured to control the sensor (102), the thermal chamber (104), and the black body radiator (106), and acquire a plurality of image data samples at a pre-defined time interval from the generated image. In an embodiment, the acquisition module (108) includes two first memories (110), the second memory (112), the logic module (114), and the delay module (116). Referring to the Figure 3, the two first memories (110) include non-volatile flash memories (110a, 110b), the second memory (112) is a volatile memory, the logic module (114) includes a logic AND gate (112), and the delay module (116) includes a delay element (116).
[0060] The non-volatile flash memory (110a) contains a table of temperature values for configuring the thermal chamber (104). The non-volatile flash memory (110b) contains a table of temperature values for configuring the black body radiator101. The volatile memory (112) works as a buffer to store data acquired from the sensor (102). In an embodiment, the acquisition module (108) reads an instantaneous temperature value from a temperature table stored in the non-volatile memory (110a) and sends a configuration command to the thermal chamber (104) to configure it to the read temperature. The acquisition module (108) then reads an instantaneous temperature value from the temperature table stored in the non-volatile memory (110b) and sends a configuration command to black body radiator (106) to configure it to the read temperature. The thermal chamber (104) and the black body radiator (106) on receiving the configuration command from the acquisition module (108), resets their respective temperatures to the read temperature values and send acknowledgement signals to the acquisition module (108). These acknowledgement signals from the black body radiator (106) and the thermal chamber (104) are latched to the input of the logic AND gate (114). The output of the logic AND gate (114) is latched to the delay element (116). On receiving acknowledgment from the thermal chamber (104) and the black body radiator (106), the logic AND gate (114) generates an output logic state. This logic state is delayed by the delay element (116). The delay element (116) provides delay so that the temperature of the sensor (102) could stabilize to a configured value. When the temperature of the sensor (102) is stabilized, the delay element (116) forwards an input trigger to the sensor (102). On receiving the trigger, the sensor (102) starts sending image data samples captured by the sensor (102). These image data samples sent from the sensor (102) are received by the acquisition module (108) buffered in the volatile memory (112). After successfully receiving the image data samples from the sensor (102), the acquisition module (108) writes these image data samples to the non-volatile memory (110).
[0061] Figure 4 illustrates a schematic diagram (400) depicting a processing module (118) of the Figure 1, according to an exemplary implementation of the present invention.
[0062] In an embodiment, the processing module (400) includes the counter (120), the step integrator (122), the divider (124), the regression module (126), and the noise parameter estimation module (128). The counter (120) counts over number of the data samples present in an image. These counts are sent from the counter (120) to the step integrator (122) and the noise parameter estimation module (128). The maximum of these counts is sent from the counter (120) to the divider (124). The step integrator (122) reads one image data sample from the non-volatile memory (110) on every count and generates one integrated image data sample on every count using Equation 1, where x is image data and xi is step integrated image data.

(1)
[0063]
[0064] The step integrator (122) writes the integrated image data samples in the volatile memory (402a). The noise parameter estimation module (128) performs the noise parameter estimation at a block (404) and reads a set of integrated image data samples from the volatile memory (402a) on every count of the counter (120) and generates a noise parameter at the last count of the counter (120). The noise parameter estimation module (128) then writes the noise parameter estimate in the volatile memory (402b). The divider (124) reads the last integrated image data sample from the volatile memory (402a) on the last count of the counter (120), and produces an output data sample. The divider (124) writes the output data sample to the volatile memory (402b). The processing module (118) also writes the image data samples from the non-volatile memory (110) to the volatile memory (402b). The regression module (126) reads image data samples, the estimated noise parameter, and the output of the divider (124) from the volatile memory (402b) and performs the regression analysis on these data samples. The regression module (126) performs the regression analysis to estimate parameters of a model described in Equation 2, where x is image data, xi is step integrated image data, zmed is noise parameter and a, b, c, d are model parameters (i.e., calibration parameters).

[0065] The regression module (126) then writes the estimated parameters a, b, c, d to the non-volatile memory (110). These estimated parameters are then loaded onto the sensor's on-board electronics and used to compute corrected image data in real-time using Equation 3, where x is image data, is corrected image data, zmed is noise parameter and a, b, c, d are model parameters (i.e., calibration parameters).

(3)

[0066] Figure 5 illustrates a schematic diagram (500) depicting noise parameter estimation, according to an exemplary implementation of the present invention.
[0067] In an embodiment, the noise parameter estimation module (128) includes the subtractor module (130), the squaring module (132), the accumulator module (134), and the running mean computing module (136). In an embodiment, the noise parameter estimation module (128) reads a set of image data samples from the volatile memory (110) on every count of the counter (120). These image data samples are then passed through the subtraction module (130) that generates an output using Equation 4, where xi is step integrated image data.

(4)

[0068] The output of subtraction module (130) is squared in the squaring module (132) and sent to the accumulator module (134). The accumulator module (134) accumulates the incoming image data samples and generates an accumulated output using Equation 5.

(5)

[0069] The running median computing module (136) receives an input from the accumulator module (134) and updates the previously computed median. The noise parameter estimation module (128) reads a set of image data samples from the volatile memory (112) for each count of the counter (120) and processes the read data samples to update previously computed median. On the last count of the counter (120), the noise parameter estimation module (128) writes the computed median to the volatile memory (112).

[0070] Figure 6 illustrates a flow diagram (600) depicting a method for calibrating infrared sensors, according to an exemplary implementation of the present invention.

[0071] The flow diagram starts at a step (602), where the black body radiator (106) and the thermal chamber (104) are configured to specific temperatures by the acquisition module (108). At the step (604), the acquisition module (108) waits for the acknowledgment from black body radiator (106) and the thermal chamber (104). On receiving acknowledgement from the black body radiator (106) and the thermal chamber (104), the acquisition module (108) generates a trigger signal for the sensor (102). On receiving the trigger signal from the acquisition module (108), the sensor (102) starts sending image data samples to the acquisition module (108). At a step (608), the acquisition module (108) captures multiple image data samples at multiple instants of time from the sensor (102) and writes them to the non-volatile memory (606). The steps (602), (604), and (608) are repeated for a fixed number of times. Each time, the acquisition module (108) configures the thermal chamber (104) and the black body radiator (106) to a specific combination of temperature values, and writes the captured image data samples to the non-volatile memory (606). When all the combinations of the thermal chamber (104) and the black body radiator (106) temperature values are finished, the acquisition module (108) gives a trigger to the processing module (118), as shown at a step (612) based on the compared value as shown at a step (610). On receiving the trigger from the acquisition module (108), the processing module (118) reads image data samples from the non-volatile memory (606). At a step (614), the processing module (118) performs a step integration process on the image data samples read from the non-volatile memory (606). At a step (616), the noise parameter estimation module (128) of the processing module (118) estimated the noise parameter using the step integrated image data samples. At a step (618), the processing module (118) performs a regression analysis using the image data samples from the non-volatile memory (606) and the output of noise parameter estimation module (128). The output data samples from the regression step (618) are then stored in the sensor's on-board electronics. For correcting image data samples captured by the sensor (102), first the image data samples are read in a step (620). Then, a noise parameter is estimated in a step (622), which indicates the amount of noise present in the image data samples. At a step (624), the pre-computed calibration parameters (628) are read from the non-volatile memory (626) and used with the noise parameter estimation module (128) to process captured image data sample and generate corrected image data samples.
[0072] Figure 7 illustrates a flowchart (700) depicting a method for shutterless temperature-robust calibration of infrared sensors, according to an implementation of the present invention.
[0073] The flow diagram starts at a step (702), sensing, by at least one sensor, electromagnetic radiations in a wavelength band and generating an image based on the sensed radiations. In an embodiment, a sensor (102) is configured to sense electromagnetic radiations in a wavelength band and generate an image based on the sensed radiations. At a step (704), setting, by a black body radiator, a pre-determined temperature for a black body radiator surface. In an embodiment, a black body radiator (106) is configured to set a pre-determined temperature for a black body radiator surface. At a step (706), acquiring, by an acquisition module, a plurality of image data samples at a pre-defined time interval from the generated image. In an embodiment, an acquisition module (108) is configured to acquire a plurality of image data samples at a pre-defined time interval from the generated image. At a step (708), processing, by a processing module, the acquired image data samples and generating at least one calibration parameter. In an embodiment, a processing module (118) is configured to process the acquired image data samples and generating at least one calibration parameter. At a step (710), estimating, by a noise parameter estimation module of the processing module, noise parameters from the acquired image data samples by using the generated calibration parameter. In an embodiment, a noise parameter estimation module (128) of the processing module (118) is configured to estimate noise parameters from the acquired image data samples. At a step (712), correcting, by a correction module, the acquired data samples based on the estimated noise parameters and generating corrected image data samples. In an embodiment, a correction module (138) is configured to correct the acquired data samples based on the estimated noise parameters and generate corrected image data samples.
[0074] It should be noted that the description merely illustrates the principles of the present invention. It will thus be appreciated that those skilled in the art will be able to devise various arrangements that, although not explicitly described herein, embody the principles of the present invention. Furthermore, all examples recited herein are principally intended expressly to be only for explanatory purposes to help the reader in understanding the principles of the invention and the concepts contributed by the inventor(s) to furthering the art and are to be construed as being without limitation to such specifically recited examples and conditions. Moreover, all statements herein reciting principles, aspects, and embodiments of the invention, as well as specific examples thereof, are intended to encompass equivalents thereof.

,CLAIMS:We claim:
1. A method for shutterless temperature-robust calibration, the method comprising:
sensing, by at least one sensor (102), electromagnetic radiations in a wavelength band and generating an image based on the sensed radiations;
setting, by a black body radiator (106), a pre-determined temperature for a black body radiator surface;
acquiring, by an acquisition module (108), a plurality of image data samples at a pre-defined time interval from the generated image;
processing, by a processing module (110), the acquired image data samples and generating at least one calibration parameter;
estimating, by a noise parameter estimation module (128), noise parameters from the acquired image data samples; and
correcting, by a correction module (138), the acquired data samples based on the estimated noise parameters and generating corrected image data samples.

2. The method as claimed in claim 1, comprising:
storing, by two first memories (110), a plurality of temperature values for configuring the thermal chamber (104) and the black body radiator (106), respectively;
storing, by a second memory (112), the acquired image data samples;
generating, by a logic module (114), an output logic state based on the acknowledgement signals; and
delaying, by a delay module (116), the output logic state for stabilizing the temperature of the sensor (102) to a configured temperature value.

3. The method as claimed in claim 2, comprising: reading, by the acquisition module (108), the temperature values from the first memory (110) and generating and transmitting a first configuration command to the thermal chamber (104) for configuration based on the read temperature values.
4. The method as claimed in claim 2, comprising: reading, by the acquisition module (108), the temperature values from the first memory (110) and generating and transmitting a second configuration command to the black body radiator (106) for configuration based on the read temperature values.

5. The method as claimed in claim 1, wherein the step of processing the acquired data samples and generating the calibration parameter comprises:
counting, by a counter (120), a number of image data samples of the generated image;
reading, by a step integrator (122), at least one image data sample and generating an integrated image data sample;
reading, by a divider (124), the integrated image data sample and generating an output data sample; and
performing, by a regression module (126), a regression analysis on the image data samples and the output data sample and generating the calibration parameter.

6. The method as claimed in claim 1, wherein the step of estimating the noise parameter comprises:
receiving, by a subtractor module (130), the acquired image data samples;
generating, by the subtractor module (130), an output by using the integrated image sample based on the received acquired image data samples;
squaring, by a squaring module (132), the generated output and generating a squared output;
accumulating, by an accumulator module (134), the squared output and the new incoming image data samples and generating an accumulated output;
computing, by a running mean computing module (136), a median based on the accumulated output; and
estimating, by the running mean computing module (136), the noise parameters based on the computed median.

7. A system (100) for shutterless temperature-robust calibration, the system (100) comprising:
at least one sensor (102) enclosed in a thermal chamber (104) with a window, the sensor (102) configured to sense electromagnetic radiations in a wavelength band and generate an image based on the sensed radiations;
a black body radiator (106) placed facing the window of the thermal chamber (104), the black body radiator (106) configured to set a pre-determined temperature for a black body radiator surface;
an acquisition module (108) configured to cooperate with the sensor (102), the thermal chamber (104), and the black box radiator (106), the acquisition module (108) configured to acquire a plurality of image data samples at a pre-defined time interval from the generated image;
a processing module (110) configured to cooperate with the acquisition module (108), the processing module (110) configured to process the acquired image data samples and generate at least one calibration parameter, the processing module (110) comprises:
a noise parameter estimation module (128) configured to estimate noise parameters from the acquired image data samples; and
a correction module (138) configured to cooperate with the processing module (110), the correction module (138) configured to correct the acquired data samples based on the estimated noise parameters and generate corrected image data samples.

8. The system (100) as claimed in claim 7, wherein the sensor (102) is enclosed in the thermal chamber (104) in a manner such that the sensor (102) faces the window towards the black body radiator (106) and a field-of-view of the sensor (102) falls within a perimeter of the black body radiator surface.

9. The system (100) as claimed in claim 7, wherein the sensor (102) is an infrared (IR) sensor.

10. The system (100) as claimed in claim 7, wherein the acquisition module (108) comprises:
two first memories (110) configured to store a plurality of temperature values for configuring the thermal chamber (104) and the black body radiator (106), respectively;
a second memory (112) configured to store the acquired image data samples;
a logic module (114) configured to receive acknowledgement signals from the thermal chamber (104) and black body radiator (106), the logic module (114) configured to generate an output logic state based on the acknowledgement signals; and
a delay module (116) configured to delay the output logic state to stabilize the temperature of the sensor (102) to a configured temperature value.

11. The system (100) as claimed in claim 10, wherein the acquisition module (108) is configured to read the temperature values from the first memory (110) and generate and transmit a first configuration command to the thermal chamber (104) for configuration based on the read temperature values.

12. The system (100) as claimed in claim 10, wherein the acquisition module (108) is configured to read the temperature values from the first memory (110) and generate and transmit a second configuration command to the black body radiator (106) for configuration based on the read temperature values.

13. The system (100) as claimed in claim 7, wherein the processing module (110) comprises:
a counter (120) configured to count a number of image data samples of the generated image;
a step integrator (122) configured to read at least one image data sample and generate an integrated image data sample;
a divider (124) configured to read the integrated image data sample and generate an output data sample; and
a regression module (126) configured to perform a regression analysis on the image data samples and the output data sample and generate the calibration parameter; and
the noise parameter estimation module (128) configured to estimate noise parameters from the acquired image data samples.

14. The system (100) as claimed in claim 7, wherein the noise parameter estimation module (128) comprises:
a subtractor module (130) configured to receive the acquired image data samples, the subtractor module (130) configured to generate an output by using the integrated image sample;
a squaring module (132) configured to cooperate with the subtractor module (130), the squaring module (132) configured to square the generated output and generate squared output;
an accumulator module (134) configured to cooperate with the squaring module (132), the accumulator module (134) configured to accumulate the squared output and the new incoming image data samples and generate an accumulated output; and
a running mean computing module (136) configured to cooperate with the accumulator module (134), the running mean computing module (136) configured to compute a median based on the accumulated output and estimate the noise parameters based on the computed median.

15. The system (100) as claimed in claim 10, wherein the first memories (110) are non-volatile memories, and the second memory (112) is a volatile memory.

Dated this 30th day of March, 2022

For BHARAT ELECTRONICS LIMITED
(By their Agent)


D. MANOJ KUMAR (IN/PA-2110)
KRISHNA & SAURASTRI ASSOCIATES LLP