Electromagnetic radiation detection system

The invention relates to a system for detecting electromagnetic radiation, such as infrared radiation.

Electromagnetic radiation detection systems are commonly used to equip auto directors with self-guided projectiles, of missile or rocket type, drones or even thermal cameras, binoculars and glasses for night visions, telescopes and more generally any observation device based on a detection of electromagnetic radiation.

Generally, electromagnetic radiation detection systems are equipped with an electromagnetic radiation sensor of the FPA matrix type (focal plane matrix, “Focal Plan Array” in English terminology). An electromagnetic radiation sensor, called simply a sensor hereinafter, is composed of a plurality of detectors sensitive to electromagnetic radiation such as thermodetectors or photodetectors. Each photo detector transforms for example the photons resulting from electromagnetic radiation into electron-hole pairs by photoelectric effect then collects the electrons in a potential well. We then speak of filling the potential well. The number of electrons collected is proportional to the number of photons received. These sensors allow obtaining images composed of pixels, each pixel being represented by at least one value coming from at least one detector of said sensor, each value depending on a level of filling in electron of the potential wells. In order to prevent the wells from being filled with electron-hole pairs generated by the photosensitive material of the sensor (ie the substrate), these sensors must be cooled.

In addition, in order to prevent photons, called parasitic photons, emitted out of a field of observation of an electromagnetic radiation detection system, also creating unnecessary electron-hole pairs, said system comprises a cooled diaphragm. This diaphragm makes it possible to limit the number of parasitic photons reaching the sensor.

Each cooling (of the substrate and of the diaphragm) is carried out by a cryostat (“dewar” in English terminology).

Due to a dispersion effect of the individual responses of each detector of a sensor, it is essential to calibrate in gain and in offset value (“offset” in English terminology) each detector of said sensor. A gain calibration corresponding to a measurement of a slope of a curve, it is necessary to acquire at least two signals (ie two values) corresponding to two different well fillings. The offset value is taken at a well filling point. Such an acquisition series is generally called a two-point and sometimes a three-point calibration. In order for each measurement to be relevant, the infills must vary at a fixed integration time. They must therefore correspond to different scenes. In addition,

Fig. 1A represents response curves of a first and a second detector of a detector of a sensor as a function of a scene temperature.

A curve 1 represents the response curve of a first detector. A curve 2 represents the response curve of a second detector. The response curves of the first and second detectors are totally different both in slope and in point of origin. An objective of calibration is to make the response curve of each detector correspond to an ideal response curve shown in Fig. 1 A by a curve 3.

Fig. 1B represents a first calibration step during which the slope of the response curve of each detector is corrected. A function making it possible to determine a gain value to be applied to the response curve as a function of a scene temperature value is then determined for the first and the second detectors. As we represent in Fig. 1B, an application of this function to curves 1 and 2 makes it possible to straighten these curves so as to obtain curves having a slope identical to curve 3.

Fig. 1C represents a second calibration step during which the shift value of the response curve of each detector is corrected.

The offset value of each curve is determined so as to obtain a response curve coincident with the ideal curve for each detector.

Conventionally, the gains are calculated once in the factory while the offset values ​​are calculated in product (that is to say in the field of use of the electromagnetic radiation detection system). However, many FPA matrices, for example those composed of MCT (Mercury Cadmium Tellurium), exhibit

instabilities of gain. Consequently, for a detector of a sensor, the gain correction applied is erroneous and the correction in offset value cannot compensate for this error. The detector concerned is then incorrectly corrected. These bad corrections can result in an image produced by the sensor, appearances of pixels, called atypical pixels, inconsistent with the other pixels of the image. These gain instabilities are all the more troublesome as the affected pixels vary from one use to another (ie from one start-up to another) of the electromagnetic radiation detection systems. In addition, electromagnetic radiation detection systems having a relatively long lifespan (for example from "15" to "20" years),

In order to identify these atypical pixels, several strategies are possible. In the factory and by “anticipation”, so-called “hardened” calibration methods carry out successive calibrations in order to increase a probability of detecting the pixels whose gain varies or is likely to vary in the future.

To manage sensor aging, some methods attempt to demonstrate the stability of sensor gains over the lifetime of the electromagnetic radiation detection system. Another solution consists in implementing calibrations during the lifetime of the device for detecting electromagnetic radiation.

These calibration methods are long and expensive. Moreover, they need to be implemented on a test bench, which means that in the event of regular calibration, the electromagnetic radiation detection system must be routed and immobilized regularly in the factory.

However, on-board calibration methods, known as active methods, are gradually developing. The main advantage of these methods is that they do not require a return to the factory for their implementation. For example, there is known a method for the a posteriori detection of atypical pixels by analysis of the offset values ​​and another using a two-point calibration device using black bodies embedded in the electromagnetic radiation detection system and allowing the calculation of gains in product. . The existing active methods significantly complicate the optical system and must generally use a Peltier module to achieve thermoelectric cooling. Using a Peltier module leads to increased power consumption and space requirements in the

the electromagnetic radiation detection system due to the volume occupied by the Peltier module. It should also be noted that these active methods cannot be (except in exceptional cases) external, which can lead to erroneous gain calculations.

Passive and active calibration methods are generally based on detecting atypical pixels by observing the gains and / or offset values ​​of the corresponding detectors. A large number of atypical pixel detection methods exist. These methods make it possible to identify a large part of the atypical pixels but generally do not make it possible to identify all of them. It is also frequent that these methods declare "for security" (ie over-declare) valid pixels as atypical pixels.

It is desirable to overcome these drawbacks of the state of the art. It is in particular desirable to provide an electromagnetic radiation detection system incorporating means allowing the implementation of an active calibration method which operates without a Peltier module, which does not over-declare atypical pixels, and which makes it possible to avoid the implementation of a hardened calibration. It is further desirable that the electromagnetic radiation detection system allow two point calibration.

According to a first aspect of the present invention, the present invention relates to an electromagnetic radiation detection system comprising: a housing defining an enclosure in which there is a partial vacuum comprising a window transparent to said electromagnetic radiation; a cold finger having a side wall closed at one end by an end wall located in line with the window; a sensor, mounted on the end wall, having a flat upper surface disposed opposite the window comprising detectors sensitive to electromagnetic radiation and cooled by the cold finger, said sensor defining an optical axis perpendicular to the flat upper surface and centered in relation to to this one; a cold screen surrounding the sensor, substantially in the shape of a dome, mounted on the cold finger and rotated around the optical axis, and comprising an upper end, disposed between the window and the sensor defining a circular shaped diaphragm centered on the optical axis and a side wall connecting a base of the cold screen at the upper end having an internal face with concavity turned towards the optical axis. The system comprises: at least one band-pass electromagnetic radiation filter having a predefined transmission coefficient, each filter being mobile and being able to assume a first position in which it is placed outside the casing facing the window and a second position in which it is placed so as not to filter any electromagnetic radiation received by the system, in said first position, each filter has in section in any secant plane containing the optical axis a concave shape facing the sensor having a profile based on a conical and / or aspherical and reflects the focal plane inside the housing; and, processing means making it possible to evaluate for each detector of the sensor a gain and an offset value by using a first value supplied by said detector when each filter is in the second position and at least a second value supplied by said detector when 'one of the at least one filter is in the first position.

The use of at least one filter in the electromagnetic radiation detection system makes it possible to simply carry out at least one two-point calibration of each detector of the sensor. This system then does not include a Peltier module.

According to one embodiment, at least a first filter among the at least one filter has a surface of which any section by a plane containing the optical axis is in the form of an ellipse or of a circle truncated by a plane perpendicular to the optical axis. and of revolution around the optical axis.

According to one embodiment, at least one of the at least one filter reflects the focal plane inside the cold screen.

According to one embodiment, each focal point of the ellipse of the section in the form of a truncated ellipse or of the circle of the section in the form of a truncated circle of the first filter is placed on the edge of the window.

According to one embodiment, the housing comprises an internal surface reflecting electromagnetic radiation and the cold screen comprises an external surface absorbing electromagnetic radiation.

According to one embodiment, each focal point of the ellipse of the section in the form of a truncated ellipse or of the circle of the section in the form of a truncated circle of the first filter is placed on the edge of the diaphragm.

According to a second aspect of the invention, the invention relates to a method for calibrating detectors sensitive to electromagnetic radiation of a sensor implemented by the system for detecting electromagnetic radiation according to the first aspect. The method comprises: positioning a filter among the at least one filter in the first position; triggering an acquisition of a first matrix of values ​​by the sensor, each value of the matrix of values ​​coming from a detector of the IR sensor; positioning said filter in the second position; trigger a

acquisition of a second matrix of values ​​by the sensor; determining for each detector of the sensor, a gain and an offset value to be applied to the values ​​coming from said detector, each determination of a gain and of an offset value of a detector uses a value from the first matrix corresponding to said detector and a value of the second matrix corresponding to said detector.

According to a third aspect of the invention, the invention relates to a computer program, comprising instructions for implementing, by a device, the method according to the second aspect, when said program is executed by a processor of said device.

According to a fourth aspect of the invention, the invention relates to storage means, storing a computer program comprising instructions for implementing, by a device, the method according to the second aspect, when said program is executed by a device. processor of said device.

The characteristics of the invention mentioned above, as well as others, will emerge more clearly on reading the following description of an exemplary embodiment, said description being given in relation to the accompanying drawings, among which:

- Fig. 1A represents response curves of a first and a second detector of a sensor as a function of a scene temperature;

- Fig. 1B represents a first calibration step during which the slope of the response curve of each detector is corrected;

- Fig. 1C represents a second calibration step during which the shift value of the response curve of each detector is corrected;

- Fig. 2 shows a sectional view of an infrared radiation detection system known from the prior art;

- Fig. 3 shows a sectional view of an infrared radiation detection system according to the invention;

- Fig. 4 shows a sectional view of part of the infrared radiation detection system according to the invention;

- Fig. 5 schematically illustrates an example of a hardware architecture of a data processing module originating from an IR sensor; and,

- Fig. 6 represents a calibration method according to the invention.

The invention is described below in the context of an infrared radiation (IR) detection system. The invention however applies to any system of

detection of electromagnetic radiation and for radiation other than infrared radiation.

Fig. 2 shows a sectional view of an infrared radiation detection system known from the prior art.

An infrared radiation detection system 2, referred to as an IR system hereinafter, comprises a housing 20 defining an enclosure in which there is a partial vacuum (approximately 10 -6 bars), provided with a window 201 transparent to IR radiation.

The IR system 2 comprises a cold finger 202 comprising a cryostat 203 (“Dewar” in English terminology) suitable for receiving a heat exchanger 213. The cold finger 202 has a side wall 204 closed at one end by an end wall 208 located to the right of window 201.

An infrared sensor 207, called IR sensor hereafter, is mounted on the end wall 208 so that it can be struck by IR radiation passing through the window 201, while being cooled by the cold finger 202. The IR sensor 207 has a flat upper surface 206, rectangular or circular, placed opposite the window 201 and made up of a matrix of detectors sensitive to IR radiation. The IR sensor 207 defines an optical axis ^ perpendicular to the planar upper surface 206 and centered with respect thereto.

In many applications, IR systems are used to observe at least one object in a scene. Said scene can be divided into two zones: a first zone comprising the object, and a second zone comprising everything that is not part of the object which we call the background. The object and the background generally have close temperatures, around "300 ° K" and therefore generate similar IR radiation. Under these conditions, it may be difficult to distinguish the object from the background. To overcome this problem, IR systems generally have a limited field of view so as to limit the radiation perceived by the sensor from the background of the object observed. As we saw above,

The IR system 1 therefore comprises a cold screen 212 substantially in the form of a dome surrounding the IR sensor 207, mounted on the cold finger 202 and of revolution around the optical axis X of the IR sensor 207. The cold screen 212 is intended for limit the IR radiation likely to reach the IR sensor 207. The cold screen 212 has a base 205, by which the cold screen 212 is mounted on the cryostat 203. The cold screen 212 also includes an upper end 210, disposed between the window 201 and the IR sensor 207 defining a diaphragm 211 of circular shape centered on the optical axis X. The cold screen 212 also comprises a side wall 209 connecting the base 205 to the upper end 210. The side wall 209 has an internal face with concavity turned towards the optical axis X.

The IR system 2 further comprises a lens system 30 adapted to focus IR radiation emanating from an object observed on the IR sensor 207. The lens system 30 may include a plurality of optical elements. In one embodiment, the lens system 30 includes a first front lens 300 and a second intermediate lens 301. Each lens is transmissive for a set of electromagnetic wavelengths corresponding to an infrared band of interest to the system. IR 2. The lens system 30 is associated with a focal plane perpendicular to the optical axis. The IR sensor 207 is located in the focal plane associated with the lens system 30.

Note that the lens system 30 is external to the cryostat and is therefore not cooled.

Furthermore, the IR system 3 comprises a processing module 213 which we will describe below in relation to FIG. 5. The processing module 213 receives a matrix of values ​​from the IR sensor 207, each value coming from a detector of the IR sensor 207. The processing module 213 applies processing to each of the values ​​received in order to generate an image from it. of the matrix of values. The processing module 213 applies in particular a gain and a predefined offset value to each of the values ​​of the matrix of values.

Fig. 5 schematically illustrates an example of a hardware architecture of a module for processing values ​​213 originating from the IR sensor 207.

According to the example of hardware architecture shown in FIG. 5, the processing module 213 then comprises, connected by a communication bus 2130: a processor or CPU (“Central Processing Unit” in English) 2131; a random access memory RAM (“Random Access Memory” in English) 2132; a ROM (“Read Only Memory”) 2133; a storage unit such as a hard disk or a storage medium reader, such as an SD (“Secure Digital”) card reader 2134; at least one communication interface 2135 allowing the processing module 213 to communicate with, for example, the IR sensor 207 and an image display module, not shown.

Processor 2131 is capable of executing instructions loaded into RAM 2132 from ROM 2133, external memory (not shown), storage media (such as an SD card), or a communication network. When the IR 2 system is powered on, processor 2131 is able to read instructions from RAM 2132 and execute them. These instructions form a computer program causing the implementation, by the processor 2131, of methods for processing the values ​​coming from the detectors of the IR sensor 207. As we will describe hereinafter, the processing system 213 is in particular able to implementing a method according to the invention, described in relation to FIG. 6, making it possible to evaluate for each detector of the IR sensor 207 a gain and an offset value.

The methods implemented by the processing system 213, and in particular the method described in relation to FIG. 6, can be implemented in software form by executing a set of instructions by a programmable machine, for example a DSP (“Digital Signal Processor”) or a microcontroller, or be implemented in hardware form by a machine or a dedicated component, for example an FPGA (“Field-Programmable Gâte Array”) or an ASIC (“Application-Specific Integrated Circuit”).

Fig. 3 shows a sectional view of an infrared radiation detection system according to the invention.

The IR system 3 of FIG. 3 uses the IR system 2 of FIG. 2. Each element identical in Figs. 2 and 3 keep the same reference. New or changed items have different references.

In one embodiment, the housing 20 remains identical in all respects in FIGS. 2 and 3. On the other hand, the lens system 30 is replaced by a lens system 40. We find in the lens system 40, the front 300 and intermediate lenses 301. The lens system 40 is associated with a perpendicular focal plane. to the X optical axis in which the IR sensor 207 is located.

A major difference between the IR system 3 and the IR system 2 lies in the insertion of at least one infrared filter, called IR filter hereafter, at the inlet of the housing 20 (ie outside the housing 20 facing the window 201). As we will see later, the insertion of at least one IR filter in the IR system 3 aims to allow at least one two-point calibration. Each IR filter inserted is a band-pass filter, mobile, of given transmission T, which is moved in front of the window 201. By changing the IR filter, due to the different transmissions T of said IR filters, the fillings of the wells vary in accordance with the requirements. needs linked to a two-point calibration and make it possible to calculate the gains and offset value of each detector of the IR sensor 207 (ie by modulating the electron filling level of the potential wells of the detectors of the IR sensor 207 without changing the integration time and by including the complete optical system). Thus, the IR filter can take a first position in which it is placed outside the housing 20 facing the window 201 and a second position in which it is placed so as not to filter any electromagnetic radiation received by the IR system. 3. In the first position, the IR filter is concave in shape facing the IR sensor 207 having a conical and / or aspheric profile and reflecting the focal plane inside the housing 20 through the window 201 or , in a preferred embodiment, inside the cold screen 212 by the diaphragm 211. In terms of shape, each IR filter therefore has in section in any secant plane containing the optical axis a concave shape facing the IR sensor 207 having a profile based on a conical and / or aspherical. Each filter is moreover characterized by a predefined transmission coefficient T and can be placed on or outside an optical path of the lens system 40.

The IR system 3 comprises means for moving each IR filter. These displacement means comprise for example, for an IR filter, a motor and an arm on which is fixed the IR filter capable of positioning or not the IR filter facing the window 201.

In one embodiment, each IR filter can be inserted between the last lens of the lens system 40 (ie here the intermediate lens 201) and the window 201 at a position allowing the focal plane to be reflected at least inside the housing. 20 and preferably inside the cold screen 212.

In one embodiment, the IR system 3 comprises a single IR filter 402.

In one embodiment, the IR filter 402 has a surface of which any section by a plane containing the optical axis X is in the form of an ellipse truncated by a plane perpendicular to the optical axis X and of revolution about the axis optical.

In one embodiment, the IR filter 402 has a surface whose section through a plane containing the optical axis is in the form of a circle truncated by a plane perpendicular to the optical axis X and of revolution around the optical axis, a circle being a special case of an ellipse.

In one embodiment, the IR filter 402 can take two positions in the IR system 3: in the first position, all the IR radiation reaching the IR sensor 207 has passed through the IR filter 402. The second position corresponds to an absence of filter IR in the IR system 3. In this second position of the IR filter 402, the IR system 3 is therefore equivalent to the IR system 2 and therefore does not filter any electromagnetic radiation received by the IR system 3. The displacement of the filter alternately in two different positions enables the IR system 3 (ie the IR sensor 207) to be supplied with two different filling levels of the potential well of each detector of the IR sensor 207 at the same integration time, which is equivalent to two acquisitions ofimages at two different black body temperatures and allows the calculation of the gain and the offset value to be applied to the values ​​from each detector.

In one embodiment, when the filter is in the first position, each focal point of the ellipse of the truncated elliptical section (respectively of the circle of the truncated circle section) of IR filter 402 is placed on the edge of the window 201. Such a positioning of the foci of the ellipse of the section in the form of a truncated ellipse (respectively of the circle of the section in the form of a truncated circle) ensures that the IR filter 402 reflects the plane focal point inside the housing 20. This positioning of the focal points of the ellipse of the section in the form of a truncated ellipse is hereinafter called general positioning.

In one embodiment, when the filter is in the first position, each focal point of the ellipse of the truncated elliptical section (respectively of the circle of the truncated circle section) of IR filter 402 is placed on the edge of the diaphragm 211. Such a positioning of the foci of the ellipse of the elliptical section (respectively of the circle of the circular section) ensures that the IR filter 402 reflects the focal plane to the 'inside the cold screen 212. This positioning of the foci of the ellipse of the section in the form of a truncated ellipse is hereinafter called optimal positioning.

In one embodiment, the positioning of the foci of the ellipse of the truncated elliptical section of IR filter 402 is determined using ray tracing techniques to ensure that IR filter 402 reflects the focal plane. inside the cold screen 212 or inside the case 20.

Fig. 4 shows a simplified sectional view of part of the infrared radiation detection system according to the invention.

The common references between FIG. 3 and FIG. 4 correspond to identical elements. It is considered in FIG. 4 that the IR filter 402 is in the first position.

IR radiations 4000 to 4002 are shown in FIG. 4. Two cases are then to be distinguished.

In the first case, the foci of the elliptical section are in the optimal position. In this case, all the IR radiation coming from the lens system 30 and passing through the IR filter 402 (such as the IR radiation 4001) and all the IR radiation coming from inside the housing 20 and reflected by the IR filter 402 (such as that the IR radiation 4000 emitted by the cold screen 212) converge inside the cold screen 212. It is noted that with a transmission coefficient T, T% of the IR radiations coming from the lens system 40 pass through the filter IR 402 and (100-7)% of the IR radiation coming from inside the box 20 is reflected by the IR filter 402. Likewise,% of the IR radiation from the interior of the housing 20 passes through the IR filter 402.

In a second case, the foci of the elliptical section are in the general positioning. In this case, it cannot be ensured that all the IR radiation coming from inside the housing 20 and reflected by the IR filter 402 converge inside the cold screen 212. Indeed, some IR radiation coming from the The interior of the housing 20 could be reflected by the internal surface of the housing 20 which would cause stray radiation which could reach the IR sensor 207. To avoid this, the internal surface of the reflective housing 20 and the external surface of the housing should be specified. Absorbent cold screen 212, for example an anti-reflective paint, a deposit or a suitable absorbent and / or diffusing treatment.

In one embodiment, the transmission coefficient T is equal to “50%”. In one embodiment, the IR system 3 comprises a first and a second IR filter, each associated with different transmission coefficients. The first IR filter is associated with a transmission coefficient T, for example equal to 30% and the second IR filter is associated with a transmission coefficient T 2for example equal to 70%. Like IR filter 302, the first and second IR filters can be moved to the first or second position. There are then three configurations of the IR system 3: a first configuration in which the first and the second filters are in the second position; a second configuration in which the first filter is in the first position and the second filter is in the second position; a third configuration in which the first filter is in the second position and the second filter is in the first position. These three configurations allow

then a three-point calibration. With a single filter, the two-point calibration has similar performance to the 3-point calibration, although different from the point of view of the correction of the pixels, in the sense that one of the measurements used to calculate the gain is also used to calculate the value. offset.

Fig. 6 represents an example of a calibration method according to the invention.

The method of FIG. 6 is implemented by the processing module 213 of the IR system 3. The method of FIG. 6 can be launched at any time by an operator, for example by pressing a button (not shown) of the IR system 3. When the method is implemented, the IR system 3 displays the same scene throughout the duration of the method. In the example of FIG. 6, the IR system 3 includes a single IR filter 402.

In a 601, the processing module 213 transmits a command to the means for moving the IR filter 402 so as to position the IR filter 402 in the first position.

In a step 602, the processing module 213 triggers an acquisition of a first matrix of values ​​by the IR sensor 207, each value of the matrix of values ​​coming from a detector of the IR sensor 207. Once acquired, the first matrix of values ​​is stored in the storage unit 2134 of the processing module 213.

In a step 603, the processing module 213 transmits a command to the means for moving the IR filter 402 so as to position the IR filter 402 in the second position.

In a step 604, the processing module 213 triggers an acquisition of a second matrix of values ​​by the IR sensor 207. Once acquired, the second matrix of values ​​is stored in the storage unit 2134 of the processing module 213.

In a step 605, the processing module performs a two-point calibration of each detector of the IR sensor 207 in order to determine, for each detector, a gain and an offset value to be applied to the values ​​originating from said detector. Each two-point calibration of a detector uses a value from the first matrix corresponding to said detector and a value from the second matrix corresponding to said detector.

The method described in relation to FIG. 6 is applicable when the IR system 3 comprises more than one IR filter by alternately positioning each IR filter in the

first or second position. When a filter is in the first position, every other filter is in the second position.

CLAIMS

1) Electromagnetic radiation detection system comprising: a housing (20) defining an enclosure in which there is a partial vacuum comprising a window (201) transparent to said electromagnetic radiation; a cold finger (202) having a side wall (204) closed at one end by an end wall (208) located in line with the window (201);

a sensor (207), mounted on the end wall (208), having a flat upper surface (206) disposed opposite the window (201) comprising detectors sensitive to electromagnetic radiation and cooled by the cold finger, said sensor defining a optical axis (X) perpendicular to the upper planar surface

(206) and centered relative thereto;

a cold screen (212) surrounding the sensor (207), substantially dome-shaped, mounted on the cold finger (202) and of revolution about the optical axis (X), and comprising an upper end (210), arranged between the window (201) and the sensor

(207) defining a circular diaphragm (211) centered on the optical axis (X) and a side wall (209) connecting a base (205) of the cold screen (212) to the upper end (210) having an internal face with a concavity turned towards the optical axis (X); characterized in that the system comprises:

at least one filter (402) of band-pass electromagnetic radiation having a predefined transmission coefficient, each filter being mobile and being able to take a first position in which it is placed outside the housing (20) facing the window (201 ) and a second position in which it is placed so as not to filter any electromagnetic radiation received by the system, in said first position each filter has in section in any secant plane containing the optical axis (X) a concave shape turned towards the sensor (207) having a conical and / or aspherical-based profile and reflects the focal plane within the housing (20); and,

processing means (213) for evaluating for each detector of the sensor (207) a gain and an offset value using a first value supplied by said detector when each filter is in the second position and at least a second value supplied by said detector when a filter among the at least one filter is in the first position.

2) System according to claim 1, characterized in that at least a first filter among the at least one filter has a surface of which any section by a plane containing the optical axis (X) is in the form of an ellipse or a truncated circle by a plane perpendicular to the optical axis (X) and of revolution around the optical axis (X).

3) System according to claim 1 or 2, characterized in that at least one filter among the at least one filter reflects the focal plane inside the cold screen (212).

4) System according to claim 1 or 2, characterized in that each focal point of the ellipse of the section in the form of a truncated ellipse or of the circle of the section in the form of a truncated circle of the first filter is placed on the edge of the window (201).

5) System according to claim 1, 2 or 4, characterized in that the housing (20) comprises an internal surface reflecting electromagnetic radiation and the cold screen (212) comprises an external surface absorbing electromagnetic radiation.

6) System according to claims 2 and 3, characterized in that each focal point of the ellipse of the section in the form of a truncated ellipse or of the circle of the section in the form of a truncated circle of the first filter is placed on the edge of the diaphragm (211).

7) Method of calibrating detectors sensitive to electromagnetic radiation of a sensor implemented by the electromagnetic radiation detection system according to any one of claims 1 to 6, characterized in that the method comprises:

positioning (601) one of the at least one filter in the first position; triggering (602) an acquisition of a first matrix of values ​​by the sensor (207), each value of the matrix of values ​​coming from a detector of the IR sensor (207);

positioning (603) said filter in the second position;

triggering (604) an acquisition of a second matrix of values ​​by the sensor (207);

determining (605) for each detector of the sensor (207) a gain and an offset value to be applied to the values ​​from said detector, each determination of a gain and an offset value of a detector uses a value of the first matrix corresponding to said detector and a value of the second matrix corresponding to said detector.

8) Computer program, characterized in that it comprises instructions for implementing, by a device (213), the method according to claim 7, when said program is executed by a processor of said device (213).

9) Storage means, characterized in that they store a computer program comprising instructions for implementing, by a device (213), the method according to claim 7, when said program is executed by a processor of said device (213).