FIELD OF THE INVENTION
The present invention relates to an improved grading machine and method for grading objects in a single pass based on its external and/or internal characteristics. More particularly, the present invention is related to a pulses grading machine and a method thereof. BACKGROUND OF THE INVENTION
Apulse is an annual leguminous crop yielding from one to twelve seeds of variable size, shape, and color within a single pod. Pulses area huge range of leguminous crops, including but not limited to kidney beans, lima beans, rice bean, black eyed peas (cowpeas), lentils, different types of broad beans like horse beans and others, garden peas, etc. Pulses are consumed in different varieties due to their high vitamin content, complex carbohydrate content, protein content and mineral content and cholesterol free nature. Pulses are rich sources of iron, magnesium, phosphorus, zinc and other minerals, which play a variety of roles in maintaining good health of human beings and animals. Therefore, pulses share a vital role and also considered as highly nutritious and significant commodity. Pulses show huge variation in size, shape, color, surface finishing etc. and are generally graded based on different external parameters like size, shape, color, surface properties or any other external characteristics by using color cameras. Pulses are divided into different classes based on their characteristics. Pulses are particularly vulnerable to pest and insect infestation during growth and storage. Poor storage practices can lead to an infection by different microbes affecting the pulses quality at disastrous extent. To address all types of variations of pulses there should be extensive analysis before grading considering their different characteristics due to tremendous external variation and easy vulnerability of pulses to internal biochemical changes or

microbial infections. All such differences should be considered while grading to provide a
quality grading.
[0003] There are some prior art references referring to pulses sorting and pulses grading based
on different parameters. It is a general practice of using optical sorting apparatus for pulses
sorting or grading.
[0004] Reference 1 is a US patent/application no. 5860531 titled as “Cereal separator using
size and specific gravity grading”. The reference 1 discloses an apparatus for separating a
cereal based on size, which comprises of frame, feed passage, grading device, specific gravity
grading and air flow producing device. Drawback of this invention is cereals are separated
only based on size which decreases the grading efficiency.
[0005] Reference 2 is a Chinese patent/application no. 1321553A titled as “Grain colour
grading machine”. The reference 2 discloses a feeding device, a predetermined number of
vibrator, a predetermined number of tracks, a predetermined number of sorting devices which
comprises of predetermined number of color components and jet detection. In this prior art
the grains are separated from unwanted materials and remaining grains are sorted based on
only color parameter. Hence, in this the categorization is based on color only that is not a
sufficient parameter for accurate grading to detect quality of the pulses.
[0006] Reference 3 is a US patent application no.US20100116974 A1 titled as “Seed Sensor
System and Method for Improved Seed Count and Seed Spacing”. It discloses an apparatus
for seed sensor system, which determines the position of the seed in the tube and speed of the
planter. In this prior art the, the seed position is determined only in X and Y direction in the
seed tube. Hence, in this the categorization is based on a single parameter is incapable in
performing an accurate grading. The prior art reference is further unable to track the velocity

and change in acceleration an also there is no control over the feeding and singulation of product stream, which results in lack of efficiency and precision.
[0007] Reference 4is a Chinese patent application no.CN202114018U title as “Machine vision system for grain real-time detection and grading”, discloses an apparatus for grading grain kernels comprising a power and transmission mechanism, a computer vision reorganization mechanism, a single granulation device and an automatic grading mechanism. The major drawback of this invention is that in this the grading is carried out based on external characteristic only without focusing on internal characteristics.
[0008] In addition to the aforementioned disadvantages, the existing systems or devices or methods for grading are not fully automated, accurate and reliable. These systems are more time-consuming, labor-intensive and low in efficiency and grading is mainly associated with external features like size, shape or color of pulses without focusing on internal or biochemical features.
[0009] Another major problem with existing pulses sorting systems is that pulses are inspected on a bulk basis and not inspected or analyzed on the individual basis for quality grading. The external and/or internal differences both are very difficult to analyze accurately on individual basis using existing grading machines. There is no effective, inexpensive objective system existing which can individually inspect pulses to analyze them in and out before grading them to judge their quality based on number of variations in their morphological and biochemical or internal characteristics. Since mixing of a batch of high quality pulses with a batch of poor quality pulses damage the economy at extremely large extents. Therefore, multiple and detailed inspection of pulses is required in many cases to assure their quality while grading.

[0010] Further, proper ejection and collection of the graded objects are also very important considerations. In general, the objects may over take another object due to density variations and due to this it is intricate to grade & eject objects accurately. Further, many times during ejection instead of one object or more than one object may get ejected simultaneously. In addition to this there might be some foreign particles such as dust, gravel or stones in the dump of objects. These conditions may lead to inaccurate grading of the objects. No aforementioned references of prior art provide systems or methods for addressing these issues. There is no system or method in the prior art which may ensure the proper tracking of the objects along the trajectory of the ejection channel and may ensure no data loss. [0011] Therefore, it would be desirable to provide an efficient automated pulses grading machine and method for grading pulses by analyzing them thoroughly considering variety of external and/or internal parameters for quality grading of pulses based on such extensive accurate analysis, thereby grading number of pulses in to multiple grades with high economic value.
SUMMARY OF THE INVENTION [0012] The proposed invention recognizes and addresses various disadvantages and drawbacks of the existing grading machines and grading methods. Taking in to account these problems and with the purpose to overcome it, the present invention provides an efficient automated pulses grading machine by considering different parameters such as object position, change in acceleration, overtaking of a heavy/light object with respect to a hollow/light object due to density variations and behavior of the object in the presence of turbulence to provide efficient and accurate grading.

[0013] Embodiments of the present invention provide an improved grading machine for grading objects, the machine comprises: a controller, a hopper for receiving objects, a feeding unit located below the hopper to receive objects from the hopper, wherein said feeding unit comprises of multiple feeder and multiple feed controllers to work co-operatively to control feeding rate of said objects, a singulator unit arranged with the feeding unit for singulating the each object to release, one by one, downwards and at least one optics unit to receive objects released from the singulator unit wherein each optics unit is comprised of: a plurality of programmable cameras configured to capture multiple image of each object in three dimensional space for determining the one or more external and/or internal parameters of each object, a plurality of light sources for enabling the programmable cameras for clear/focused capturing of each object, an image processing unit for processing each said captured image of each object into at least one parameter, and a first storage module configured to store each parameter. [0014] The improved grading machine further comprises one or more duct arranged to receive objects from the at least one optics unit wherein the each duct comprised of: at least one multiple sensor layers arranged at pre-determined distance in the duct for analyzing each object horizontally, a distributed network comprises at least one multiple sensor layer and multiple sensor controllers wherein there is a sensor layer controller to coordinate with respective sensor layer of corresponding duct. There is at least one network controller for controlling all sensor layer controllers of corresponding duct, wherein each sensor layer comprises of at least one sensor to continuously track one or more variable factors including real-time position of each object, change in acceleration, velocity, overtaking of a heavy/light object with respect to a hollow/light object and

behavior of the each object in the presence of turbulence; multiple ejectors arranged in the duct at different ejection points and configured to actuate on receiving signals from the controller for ejecting the objects and at least one line scan camera arranged at each ejection point to sense the presence of objects viewing the one or more duct for detecting the one or more variable factors of each object at a given point of time. [0015] A comparator of the controller is configured to determine the grade of each object and to determine ejection point of the each graded object by comparing the one or more variable factors of each object detected by the at least one sensorof multiple sensor layer and the one or more variable factors of each object detected by the line scan camera with the higher threshold value stored in a second storage module and to vary the lower threshold value according to the detected one or more variable factors of each object to determine the ejection points for ejecting the objects from the multiple outlets. Further multiple collecting locations are configured to collect the objects from multiple outlets based on the grade of each object.
[0016] In an aspect of the present invention, the improved grading machine provides the singulator is typically a pick and place unit that singulates the objects one by one to achieve predetermined distance between the objects to avoid multiple ejection and object over tacking with respect to another object thus facilitating effective analysis and ejection.
[0017] Due to density variations, the object over takes another objects inside the duct and also the heavy/light object over takes another particle. The present invention is capable of recognizing objects in aforementioned condition and intelligently grades and ejects them by determining the change in acceleration and object position with highest

efficiency. Even, the present invention is capable of recognizing and grading the objects in the condition where the objects are in the midst of foreign particles like grime or dust. An aspect of the present invention provides a method for grading objects based on pre-defined parameters. The method comprises the steps of: feeding the objects into hopper and passing on the objects from hopper to the feeding unit wherein feeding unit regulate the feed rate of the objects, singulating the objects by the singulator unit (102) to convey the object one by one to the optics unit for optical analysis of each singulated object and capturing and processing multiple images of each singulated object in three-dimension space to detect one or more pre-defined external and/or internal parameters of each object, storing the parameters corresponding to each object in a first storage module and further sending to the controller. The controller further grades the each object by comparing each parameter corresponding to each object stored in the first storage module with a pre-determined parameter stored in a second storage module of the controller using a comparator.
[0018] The pre-determined parameters are pre-computed to determine multiple grades of objects based external and internal parameters such as size, shape, color, surface properties, pulses profile, external defects, mechanical injuries, moisture content within pulses, rancidity level, nutrition content, maturity level of pulses, any microbial infection or any microbial activity or biochemical activity inside pulses which make pulses rotten, infected or diseased, or any other possible characteristics. [0019] The method further comprises the steps of: allowing the graded object to pass through a duct, one by one, and analyzing one or more variable factors by a plurality of sensors arranged in the duct and (or) at least one line scan camera arranged at/proximal

to each ejection point, communicating the one or more variable factors detected by the plurality of sensors through network controller and (or) the at least one line scan camera directly to the controller for determining ejection points of each object, actuating multiple ejectors for ejecting objects by determining the ejection points for the objects by analyzing the one or more variable factors and collecting the objects in multiple collecting locations of pre-defined grade from multiple outlets configured opposite to the multiple ejectors.
[0020] Some of all aforementioned advantage of the present invention are considered by providing a system or method to grade pulses or similar objects whose properties are similar to pulses based on internal and external parameters such as size, shape, color, hollow or broken pulses etc.
[0021] A further advantage of present invention is accrued by providing object trajectory tracking mechanism to track the change in acceleration and position the objects when they made fall in duct by singulator unit and till their reaching on the collecting units. Further, it is also advantage over multiple prior-arts to provide a singulator unit for maintaining a predetermined distance the objects for easy ejection. [0022] Above all the main advantage of the present invention is facilitated by the master controller. Due to density variations, generally, one object overtakes other one in duct, in that case the master controller is intelligent to track and eject such object accurately. The controller considers parameters inside the duct such as object position, change in acceleration while the next or previous object is ejected. The effect of turbulence created by ejection of the next or the previous object is also compensated by

the controller. By evaluating on these parameters, the ejection of object can be timed
accurately, resulting in a very accurate ejection.
[0023] Accordingly, few objects of the present invention are listed below:
It is an object of the present invention to provide an improved pulses grading machine
and a method there of, where change in acceleration with respect to object is tracked
along the trajectory path for efficient grading.
It is another object of the present invention is to providean improved pulses grading
machine comprising a singulator unit by which the objects are picked and placed one by
one to maintain the pre-determined distance between the objects. This singulation
mechanism helps in easy analysis and ejection of every single object.
It is a further object of the said invention is that the objects are continuously being
tracked, when one object overtakes the other due to density variations.
It is a yet an object of this invention is to provide a master controller that considers
different parameters before applying the fluid for ejection such as current position of the
object, change in acceleration, velocity, behavior of the object in presence of turbulence
inside the tube, which increases accuracy of the system.
It is still another object of the present invention to find the change in acceleration and
position of the object in order to maintain accuracy.
It is another object of the present invention to provide an improved multi-vision and
multi spectral pulses grading machine and a method for grading pulses into multiple
grades in a single pass based on external or biochemical/internal parameters, wherein
the analysis of each & every object performed with higher precision.

This invention is pointed out particularly with the appended claims. Additional features
and the advantages of the improved grading machine will become apparent to those
skilled in the art by referring to the following detailed description taken in conjunction
with the accompanying drawings.
BRIEF DISCRIPTION OF THE DRAWING
[0024] The above mentioned and other objects, features, and advantages of the
invention will best be understood from the following description of various
embodiments thereof, selected for the purposes of illustration, and shown in the
accompanying drawings. The drawing illustrates various embodiment of the invention
which is not to be construed as limited.
FIG. 1a is a schematic diagram illustrating the arrangement of non-limiting elements of
an improved grading machine and FIG. 1b is an illustration of high speed line scan
camera of in accordance to an embodiment of the present invention;
FIG. 2a is an illustration of overlapping of objects in duct and FIG.2b is wave form
representation of overlapping of objects, in accordance to an embodiment of the present
invention;
FIG.3 is an illustration of minimum distance between the object in duct and FIG.3b. is
wave form representation of minimum distance between the objects;
FIG .4 is an illustration of dust or other noise in the duct;
FIG.5 is an illustration of Turbulence in duct; and
FIG.6 is an illustration of overtaking of object with respect to the one based on density
variations.

DETAILED DESCRIPTION OF THE INVENTION [0025] The present invention will now be described in a great detailed manner with reference to the accompanying exemplary drawings for the purposes of illustrating non-limiting embodiments of the present invention.
As used herein, the term ‘object’ refers to a feed for the purposes of the invention and the feed may be any pulses or similar objects whose properties are similar to pulses, which may be ‘wet’ or ‘dried’ or ‘whole’ or ‘broken’ or ‘split’ or ‘husked’ or ‘un husked” or ‘mature’ or ‘immature’ or combination or mixtures of different types of pulses also objects which are naturally occurring or synthetically manufactured having a statistical distribution of objects including variations in density of the same having an object size in the range of 1-100 mm.
As used herein, the term ‘external parameters’ shall refer to any morphological, extrinsic parameters related to pulses which may include but not limited to size, shape, color, surface properties, pulses profile, or alike any other properties, external defects, mechanical injuries, or any other possible characteristics.
As used herein, the term ‘internal parameters’ shall refer to any intrinsic or biochemical characteristics related to pulses which may include but not limited to pulses physical properties like, moisture content within pulses, rancidity level, protein content, maturity level of pulses, any microbial infection or any microbial activity or biochemical activity inside pulses which make pulses rotten, infected or diseased, or any other possible characteristics.

As used herein, the term ‘camera’ shall refer to the programmable cameras which are
programmed as per the need of the invention and these cameras can be multiple in
number. The cameras can be “regular color cameras” or “multi-spectral cameras”.
As used herein, the term ‘multi-spectral cameras’ work at different frequencies of
electromagnetic spectrum (multi-spectrum) like visible, ultra-violet, infra-red, x-ray etc.
for enhanced analysis of the pulses based on external and/or internal parameters of the
pulses or any other object of interest.
As used herein ‘pre-determined distance’ is the minimum distance achieved by
singulation.
As used herein ‘pre-determined parameters’ are pre-computed to determine multiple
grades of objects based on the size, shape, color, surface properties, pulses profile,
external defects, mechanical injuries, moisture content within pulses, rancidity level,
nutrition content, maturity level of pulses, any internal or microbial infection or any
microbial activity or biochemical activity inside pulses which make pulses rotten,
infected or diseased, or any other possible characteristics.
[0026] Embodiments of the present invention disclose an apparatus to grade pulses in
to multiple grades based on external and/or internal characteristics.
Referring to FIG. 1, in an embodiment of the present invention, an improved grading
machine (10) includes controller (Not shown in figures) that controls the functionality
of entire grading machine (10). The non-limiting elements of the improved grading
machine (10) are a hopper (100), feeding unit (101) which comprises of one or more
feed sensors and one or more feed controllers; singulator unit (102), optics unit (103),
multiple outlets [shown as P1, P2, P3….Pn] (106) , at least one duct (109)which

comprises a plurality of sensors (104) arranged in corresponding sensor layers, ejectors (108) arranged at the ejection points [shown as T1, T2, T3…..Tn] and multiple collection location [shown as C1, C2, C3…….Cn] (107); and at least one high speed
line scan camera (110) [shown as L1,L2, Ln] arrange at/proximal to the ejection
points [shown as T1, T2, T3…..Tn]. As used here ‘n’ is any natural number . [0027] The duct (109) includes of a plurality of sensors (104) arranged in corresponding sensor layers along the trajectory of the falling object (105) to detect one or more variable factors of each falling object (105) in real-time. The one or more variable factors include position of objects, change in acceleration, velocity, overtaking of a heavy/light object with respect to a hollow/light object based on density variations and behavior of the each object in the presence of turbulence to achieve accurate and efficient grading.
[0028] The ejectors (108) are designed to blow vacuum/air/fluid from opposite sides of the multiple outlets multiple outlets (106) [shown as P1, P2, P3….Pn] to achieve efficient ejection. The at least one high speed line scan camera (110) [shown as
L1,L2, Ln] is installed at each ejection point to sense the presence of object viewing
single and/or multiple ducts at a given point of time. The high speed line scan camera
can detect the presence of multiple objects at multiple ducts at a time.
Hopper
[0029] Pulses having different external or internal characteristics are fed into the
hopper (100). The hopper (100) is configured to feed objects of having different internal
and external characteristics. The hopper (100) acts a reservoir for providing objects to
the feeding unit (101). It can be structured as an inverted pyramidal cone to receive an

amount of the objects. Non-restraining to the shape of hopper (100), in other embodiments the hopper (100) can be structured to have any shape for fulfilling its purpose of feeding such as hollow cylindrical shape or rectangular shape etc. Feeding unit
[0030] In an embodiment of the present invention the feeding unit (101) may be a vibratory unit that includes: a slider, at least one set of vibrators comprising of a first vibrator and a second vibrator. The second vibrator is inverted vibrator. The feeding unit (101) further includes multiple feeding sensors and a feeding controller. The feeding unit (101) is vibratory and can have some other modifications for maintaining a specific queue system to accommodate varied sized and shaped pulses as a feed in a systematic way. The multiple feeding sensors and feeding controllers work co¬operatively for sensing the feed rate and efficiency of feeding and further regulate the flow of objects.
[0031] In another embodiment of present invention, there can be multiple feeding units. Each feeding unit includes at least one feed controller for one feeder to control bulk flow of objects from corresponding feeder into corresponding singulator unit (102) and optics unit (103) and further into corresponding duct (109) for effective feeding, thereby making the feeding unit (101) as completely automated and controlled based on the need of number of objects to be fallen in particular duct. Feed controller of corresponding feeder receives signals related to flow of objects from a network controller of sensor network of corresponding duct (109) through controller as the feed controller of corresponding duct is coupled with the controller for effective feeding of

objects from corresponding feeder into corresponding optics unit (103) and further into corresponding duct (109). Singulator unit
[0032] This invention includes a singulator unit (102) to singulate the objects one by one. The singulator unit (102) singulates the pulses one by one and maintains the predetermined distance between the objects to avoid overlapping and over tacking of the objects with respect to another object thus facilitating effective analysis and ejection. Optics unit
[0033] In one embodiment, the optics unit (103) includes a plurality of programmable cameras, multiple light sources, an image processing unit and a first storage module. The programmable cameras capture multiple images of the objects under inspection in three-dimensional space. The captured image is without reconstructing the plurality of images of object, this feature helps in accurate and statistical analysis of each object under inspection. The images are captured to determine at least one external and/or internal parameters of each object under Inspection. The captured images are stored in the first storage module.
[0034] The image processing unit processes each of the captured images of each object and converts the images into at least one parameter. Typically, the external parameters are the external features like at least one of size, shape, color, surface properties and object profile and typically, the internal parameters are internal features like moisture content, biochemical changes or microbial infections. However, the present invention is not limited to the above mentioned parameters and any other

parameters can also be considered for inspection. The first storage module stores each parameter generated in the image processing unit. Duct
[0035] In an embodiment of the present invention, the one or more duct (109) arranged to receive objects from the at least one optics unit (103). The duct (109) includes a distributed network comprising multiple sensor layers which are arranged from the starting point of each duct to track objects.
[0036] The one/multiple sensor layers include multiple sensor layer controllers to coordinate with the corresponding multiple sensor layers. Each sensor layer controller is provided to coordinate with respective sensor layer of the corresponding duct. Further, a network controller is provided for controlling all sensor layer controllers of corresponding duct. Each sensor layer includes of at least one sensor (104) and sensor controller to continuously track one or more variable factors including real-time position of each object, change in acceleration, velocity, overtaking of a heavy/light object with respect to a hollow/light object and behavior of the each object in the presence of turbulence. Sensors sense the object along the trajectory such as position of the object change in acceleration, velocity and behavior of the object in the presence of turbulence these parameters are send to master controller.
[0037] In an embodiment of the invention, the sensor (104) tracks each object and sends signal the sensor controller, further the sensor controller sends signals to the sensor layer controller of each sensor layer which communicates with the network controller. Finally, network controller sends signals to the controller.

Further, at least one high speed line scan camera (110) [shown as L1,L2, Ln] is
installed at each ejection point to sense the presence of object viewing single and/or
multiple ducts at a given point of time. The high speed line scan camera can detect the
presence of multiple objects at multiple ducts at a time.
[0038] In an embodiment of the present invention, the sensors (104) are located in the
duct (109) and line scan camera (110) is located at some pre-defined distance from the
duct (109). In this embodiment, only rays from line scan cameras (110) are passed in the
duct (109) through at least one hole made in the duct (109). Line scan cameras (110) are
placed in proximity to duct (109) not inside duct (109).
[0039] The plurality of sensors (104) and high speed line scan camera (110)
[L1,L2, Ln] detect the different parameters like object position, velocity, change in
acceleration, behavior of the object in the presence of turbulence, tracking of heavy/light object overtakes light/hollow object at various ejection points and at the ejection points, respectively. As at ejection point there are more chances of air variation and turbulence due to the ejection of other objects, so it is significant to track each object at ejection point horizontally for same parameters which the sensors (104) have already tracked. Due to the vertical and horizontal tracking of each object, there is no data loss in the improved grading machine (10), therefore, objects can be graded and ejected accurately.
[0040] In an embodiment of the present invention, the sensor (104) tracks the objects and sends real time signals related to object position, change in acceleration, velocity or density variations continuously to the controller through the network controller. The

line scan camera (110) tracks the objects particularly at the ejection point and sends real time signals directly to the controller.
[0041] The multiple sensor controller receive the one or more variable factors corresponding to the each object such as change in acceleration, velocity, object position, and behavior of the object in the presence of turbulence are sent to the controller through the network controller. Further considering these parameters, which are sent by the network controller of the multiple sensor layer controllers to the controller, the controller decides the grade and the ejection point for each object. Further, the controller compensates turbulence created by ejection of the next or the previous object. Due to the turbulence the object may follow undesired path because of unsteady movement of air or any fluid inside duct.
[0042] The controller takes into consideration of object tracking data provided by the sensor (104) and (or) high speed line scan camera (110) to decide to eject the object or not. This enable the present grading machine to attain efficiency of the ejection is far superior to the existing machines or inventions.
[0043] The object may over take another object inside the duct due to density variations and if it happens, it would be difficult to grade & eject accurately; instead of one object or more than one object may get ejected simultaneously. Further, if the heavy\light object over takes another particle, it is tracked along the duct to avoid multiple ejections at a time. Even in the condition where the objects are over taking each other, the invention can easily grade and eject them by determining the change in acceleration and object position with highest efficiency.

[0045] In one embodiment, the sensors of multiple sensor layers can be IR, ultrasonic, proximity sensors or combination thereof/ without having any restriction/limitation in the invention. Controller
[0046] In one embodiment of the present invention, the controller is one master controller. In another embodiment, the controller may be more than one controller. The controller includes a comparator and a second storage module. The controller can decide the accurate position of grade of each object to reach to its grading point in real time. The information about position, velocity etc. of each conveying object is analyzed by all sensors and the sensor controllers of corresponding duct accurately in real time as sensor controllers are always active during the grading process to receive signals from one or multiple sensor layered upright from corresponding duct (109) to sense each grade which can randomly come across any sensor of corresponding duct (109). Collecting locations
[0047] The improved grading machine (10) includes multiple collecting location [shown as C1, C2, C3…….Cn] (107)arranged to receive objects according to their grades from the multiple outlets (106) [shown as P1, P2, P3….Pn]. In an embodiment of the invention, there can be “n” numbers of depending of the 1 to “n” grades of the objects to collect the objects of each grade separately. The collecting bins (107) according to their different parameters like size, shape & color or any other internal or external characteristics. The multiple collecting location [shown as C1, C2, C3…….Cn] (107)can have bins arrange nearly or remotely to store graded objects.

[0048] Referring to FIG. 1b, at least one line scan camera (110) in proximity of ejection points is placed. The sensor (104) detects one or more variable factors of each falling object. The duct (109) further includes one/more sensor controller that is configured to control/command the one/multiple sensor (104). By the mean of the sensors (104), the objects are tracked and graded inside duct along the trajectory of the object. Further, the line scan camera (110) detects the objects to be ejected. The high speed line scan camera (110) senses the different parameters like object position, velocity, change in acceleration, behavior of the object in the presence of turbulence, tracking of heavy/light object overtakes light/hollow object. The sensed data is first sent to the each sensor controller and further the sensor controller sends data to the network controller which further sends data to the controller. The controller takes into consideration of the sensor data and (or) high speed line scan camera data to decide to eject the object into the bins or not. By this the efficiency of the ejection is far superior to the existing machines or inventions.
[0049] In an embodiment of the present invention, the numbers of high speed line scan cameras (110) are not limited to the invention. There can be ‘n’ no. of line scan cameras arranged with one or more ejectors (108).
[0050] In an embodiment of the present invention, the line scan cameras (110) are positioned outside duct in proximity of multiple ducts in such a way that at each ejection point of all/ multiple duct has one line scan camera (110) to track the objects. The controller considers variable factors of each falling object such as object position, change in acceleration, velocity, behavior of object in turbulence and the conditions

where the object over takes another object inside duct based on the density variations. All sensor controllers send signals to controller through the network controller. [0051] FIG. 2a is a illustration of overlaping of objects in duct (109) and FIG.2b is wave form reprsentation of overlapping of objects. Refering to the figures, the object1(201) and object2(202) are overlapped along the tranjectory path, at this moment it is difficult to track the individual object and due to overlapping. The plurality of sensor (104) will consider this as one object as shown in the the representation of FIG.2b.
[0052] In FIG. 2b, ‘ T1’ is the lowest threshold value and ‘T2’is the highest threshold value. The X can be any substantial property such as voltage, current or any other parameters and t is time. The signals are respresented time versus voltage/current/other parameters in x direction and y direction respectively. The objects are not differentiated because signal is partially above (203) the lower threshold value ‘T2’. Due to no separation between the objects, the sensor considers this stand point as one object. The invention overcomes this issue by maintaining the predetermined distance between the objects and the controller shifts the lower threshold ‘T1’ in real time towards higest threshold ‘T2’ by considering different parameters of objects sent by the network controller of the multiple sensor layers of each duct from which objects are differentiated individually and accurately.
[0053] FIG. 3a is an illustration of minimum distance between the objects in the duct (109) and FIG. 3b is wave form representation of minimum distance between the objects. Reffering to FIG. 3a the objects are moving at predetermined distance, from this individual objects can be identified separately. This is achieved by singulator

mechanism,which helps in easy grading and efficient ejection. In an embodiment of the present invention the, overtaking of a heavy/light object and overlapping of two or more objects are determined by comparing the one or more substantial property of objects to a pre-defined threshold value stored in the controller.
[0054] FIG.3b is a wave form representation, object1(201) and object2 (202) both are idententified separately above the higest threashold value ‘T2’and below lower threshold value ‘T1’. The objects are easly identified the sensors (104) of multiple sensor layers.
[0055] FIG. 4 is a illustration of dust or other noise in the duct. Referring to the figure, it represents the dust signal(204) which can be easily differentiated from objects.The signal lies inbetween the two threshold values ‘T1’ and ‘T2’. The signal which covers both highest threshold value ‘T2’ and lower threshold value ‘T1’ is considered as object. [0056] FIG.5 is a illustration of Turbulence (205) in duct.The object1(201) and object(202) undergo undesired path due to turbulence (205).The behaviour of the object is tracked in presence of turbulence (205) and compensated by the controller. The contorler receives the sensor inputs like current position of the object,change in acceleration,velocity and heavy/light object overtakes hollow/light object. From these parameters the controller decides grade type and eject the objects effeciently and turbulence is further compensated by monitoring the amount of fluid blow to eject the objects.
[0057] FIG.6 is a illustration of overtaking of object with respect to other one based on density.The heavy/light object(201) overtakes the hollow/light object (202) because of the density variations,due to this two or more objects may come side by side. This

issue is avoided by contineous tracking of object positon, change in acceleration and velocity along the trajectory of the object and these parameters are considerd before ejection.
[0058] In one embodiment, the improved grading machine (100) includes, a controller, a hopper (100), feeding unit (101) which comprises of one or more feed sensors and one or more feed controllers; at least one singulator unit (102), at least one optics unit (103), multiple outlets [shown as P1, P2, P3….Pn] (106) , at least one duct (109 )which comprises a plurality of sensors (104) arranged in corresponding sensor layers, ejectors (108) arranged at the ejection points, and multiple collecting location (107); and at least one high speed line scan camera (110) arranged at/proximal to the ejection points. The optic unit (103) comprises of white light unit, an ultra-violet unit, an infra-red unit, an x-ray unit or a laser ray unit and/ or combination thereof or the optics unit with any radiation comes under electromagnetic spectrum that releases appropriate rays for examining at least one external and/or internal characteristics and a processing unit that processes the appropriate rays for determining grade of the object under examination and not limiting the scope of examination to the use of the cameras and the image processing unitas described in this disclosure. Any examination unit or processes that can determine at least one external and/or internal characteristics of the object under examination shall be considered within the scope of the present invention. Embodiment of the present invention, further includes a method for grading objects by using an improved grading machine. In first step the objects are fed into a hopper (100). [0059] The objects are further directed from the hopper (100) to the feeding unit (101) wherein feeding unit regulate the feed rate of the objects. The objects fall from hopper

(100) to the feeding unit due to gravity. Further, the objects are singulatedby the singulator unit (102) to convey the object one by one to the optics unit for optical analysis of each singulated object. In next step, the each singulated object is captured by the multiple programmable cameras. The multiple programmable cameras are set at different angles which capture images in three-dimensional space. The captured images are further processed by the image process to detect one or more pre-defined external and/or internal parameters of each object. The said parameters are stored corresponding to each object in a first storage module. Further the comparator of controller compares the parameters of each object stored in the first storage module with a pre-determined parameter stored in a second storage module of the controller. In another step, the graded object are allowed to pass through a duct (109), one by one, and one or more variable factors are further analyzed verticallyby a plurality of sensors arranged in each duct. Further, the same variable factors are analyzed horizontally by the at least one line scan camera (110) arranged at each ejection point to ensure no data loss.In next step, the one or more variable factors are detected by the plurality of sensors (104) and communicated to the controller through the newtork controller. [0060] The one or more variable factors detected by the at least one sensor (104) are communicated to the controller in a way such that the network controller of the sensor network receives signals related to the one or more variable factors of each object in real time to determine the position, velocity and density of object accurately in real time from each sensor layer controllers of corresponding duct. Further, network controller of the sensor network sends the said signals to the controller as each network controller of each duct is coupled to the controller. Further, the controller receives the signals from

the network controller of the sensor network of corresponding duct by as the object cuts the multiple rays of corresponding sensor layers, thereby controller anticipates the exact position, velocity and density of each object accurately in real time before the arrival of the object at ejecting point. Further, the same parameters of each object are deatched by the at least one line scan camera (110) at an ejection point. Based on the received parameters, the controller determines the ejection point for each object and actuates multiple ejectors (108) for ejecting objects. Further, the objects are directed towards the multiple pre-determined outlets and pushed by the ejectors (108) to the multiple collecting locations (107) configured opposite to the multiple ejectors (108). [0061] In an embodiment of the present invention, the each ejector (108) of corresponding duct (109) is coupled to the controller; therefore each ejector (108) receives signals related to ejection of each object sent by the controller before the arrival of each grade of object in corresponding duct. When the particular grade of object conveys near particular grading point in corresponding duct across its trajectory at particular position, these ejectors (108) of corresponding duct opens a valve to eject a beam of pre-defined duration of high pressure air or high pressure fluid to eject the particular grade of object.The pressure applied by said ejectors eject each grade of object accurately, thereby making each grade of object to fall into the corresponding desired collecting location (107) through corresponding multiple outlets (106) for collecting different grades of objects into multiple grades in a single pass. [0062] In some other embodiments of the present invention, the improved grading machine (10) may have a plurality of each element as stated above. For example, there

can be a single duct (109) in one embodiment while in other embodiment there can be ‘n’ no. of ducts.Similary, a number of modification are possible. [0063] The aforementioned embodiments provides some or all aforementioned advantage of the present invention by providing a system or method to grade pulse or similar objects whose properties are similar to pulse based on internal and external characteristics such as size, shape, color, hollow or broken pulses etc. Further advantage of present invention is accrued by providing object trajectory tracking mechanism to track the change in acceleration and position the objects when they made fall in duct by singulator unit and till their reaching on the collecting units. Further, it is also advantage over multiple prior-arts to provide a singulator unit for maintaining a predetermined distance the objects for easy ejection. Above all the main advantage of the present invention is facilitated by the controller as a ‘master controller’. Due to density variation, generally, one object overtakes other one in duct, in that case the master controller is intelligent to track and eject such object accurately. The controller considers parameters inside the duct such as object position, change in acceleration while the next or previous object is ejected. The effect of turbulence created by ejection of the next or the previous object is also compensated by the master controller. By evaluating on these parameters, the ejection of object can be timed accurately, resulting in a very accurate ejection.
[0064] In another embodiment of the invention, the improved grading machine (10) includes , a controller, a hopper (100), feeding unit (101) which comprises of one or more feed sensors and one or more feed controllers, singulator unit (102), optics unit (103), multiple outlets (106) , at least one duct (109) which comprises a plurality of

sensors (104) arranged in corresponding sensor layers, ejectors (108) arranged at the ejection points and multiple collection location (107); and at least one high speed line scan camera (110) arrange at/proximal to the ejection points. The optic unit includes one of an ultra-violet unit, an infra-red unit, an x-ray unit or a laser ray unit that releases appropriate rays for examining at least one internal and/ or external characteristics or any other characteristics of objects and a processing unit that processes the appropriate rays for determining grade of the object under examination and not limiting the scope of examination to the use of the cameras and the image processing unit as described in this invention. Any examination unit or processes that can determine at least one internal, external or any other characteristics of the object under examination shall be considered within the scope of the present invention.
[0065] As will be readily apparent to those skilled in the art, the present invention may easily be produced in other specific forms without departing from its essential characteristics. The present embodiments are, therefore, to be considered as merely illustrative and not restrictive, the scope of the invention being indicated by the claims rather than the foregoing description, and all changes which come within therefore intended to be embraced therein.

What is claimed is:
1. An improved grading machine (10) for grading objects, the machine (10) comprises:
-at least one controller;
-at least one hopper (100) for receiving objects;
-at least one feeding unit (101) located below the hopper to receive objects from the
hopper, wherein said feeding unit comprises of multiple feeder and multiple feed
controllers to work co-operatively to control feeding rate of said objects;
- at least one singulator unit (102) arranged with the feeding unit (101) for singulating
the each object to release, one by one, downwards;
-at least one optics unit (103) to receive objects released from the singulator unit (102)
wherein each optics unit is comprised of:
■ a plurality of programmable cameras configured to capture multiple image of each object in three dimensional space for determining the one or more external and/or internal parameters of each object;
■ a plurality of light sources for enabling the programmable cameras for clear/focused capturing of each object;
■ an image processing unit for processing each said captured image of each object into at least one parameter; and
■ a first storage module configured to store each parameter;
-at least one or plurality of ducts (109) arranged to receive objects from the at least one optics unit (103) wherein the each duct (109) comprised of:
■ a distributed network comprising one/multiple sensor layers arranged at pre-determined distance, from the starting point of each duct (109) till the

last ejecting points of objects, wherein in an operative configuration,
multiple sensor controllers disposed proximal to the distributed network
coordinate with corresponding multiple sensor layers, wherein there is a
sensor layer controller to coordinate with respective sensor layer of
corresponding duct, and at least one network controller disposed proximal
to the multiple sensor controllers for controlling all sensor layer controllers
of corresponding duct to communicates with the controller and, wherein
each sensor layer comprises of at least one sensor (104) to continuously
track one or more variable factors including real-time position of each
object, change in acceleration, velocity, overtaking of a heavy/light object
with respect to a hollow/light object and behavior of the each object in the
presence of turbulence; and
■ multiple ejectors (108) arranged in the duct (109) at different ejection
points and configured to actuate on receiving signals from the controller for
ejecting the objects;
-at least one line scan camera (110) arranged at each ejection point to sense the
presence of objects viewing the one or more duct for detecting the one or more
variable factors of each object at a given point of time;
-a comparator of the controller configured to determine the grade of each object and to determine ejection point of the each graded object by comparing the one or more variable factors of each object detected by the at least one sensor (104) of multiple sensor layer and the one or more variable factors of each object detected by the line scan camera (110) with the higher threshold value stored in a second storage module

and to vary the lower threshold value according to the detected one or more variable factors of each object to determine the ejection points for ejecting the objects from the multiple outlets (106); and
-multiple collecting locations (107) arranged to collect the objects from multiple outlets (106) based on the grade of each object.
2. The improved grading machine (10) for grading objects as claimed in claim 1, wherein the singulator unit (102) singulates the objects one by one to maintain the predetermined distance between the objects to avoid overlapping during optical analysis thereby enables the controller to determine accurate individual grade of the each object.
3. The improved grading machine (10) for grading objects as claimed in claim 1, wherein the at least one sensor (104) of each sensor layer tracks object (105) vertically and the same object (105) is been tracked by the line scan camera (110) at the ejection point horizontally thereby avoiding any data loss.
4. The improved grading machine (10) for grading objects as claimed in claim 1, wherein the multiple ejectors (108) are air/vacuum/pressure operated which receive signals from the controller regarding ejection point and the amount of actuation required to eject the objects.

5. The improved grading machine (10) for grading objects as claimed in claim 1, wherein overtaking of a heavy/light object and overlapping of two or more objects are determined by comparing the one or more substantial property of objects to a pre-defined threshold value stored in the controller.
6. The improved grading machine (10) for grading objects as claimed in claim 1, wherein the controller is configured to:

-determineexact, accurate, final grade of each object by receiving signals related to individual grade of each object sent by each optics unit,
-determine accurate ejection points of each object by comparing the signals sent by each network controller of each sensor network of the at least one sensor (104) related to one or more variable factors of each objects and with signals related to one or more variable factors from at least one line scan camera; and
-determine pressure to be required to blow object at ejection point for accurate ejection of each graded object.
7. The improved grading machine (10) for grading objects of claimed 1, wherein the pre-defined external and/or internal parameters include size, shape, color, surface properties, pulses profile, or alike any other properties, external defects, mechanical injuries; and moisture content within objects, rancidity level, nutrition content, maturity level of object, any microbial infection or any microbial activity or biochemical activity inside object which make objects rotten, infected or diseased.
8. A method for grading objects by using an improved grading machine, the method comprising the steps of:
a. feeding of objects into a hopper (100) and passing on the objects from the
hopper (100) to a feeding unit (101) with regulate the feed rate of the objects;
b. singulating the objects by a singulator unit (102) to convey the object one by
one to an optics unit for optical analysis of each singulated object;
c. capturing and processing multiple images of each singulated object to detect
one or more pre-defined external and/or internal parameters of each object,

storing the parameters corresponding to each object in a first storage module and further sending to a controller;
d. grading the each object by comparing each parameter corresponding to each
object stored in the first storage module with pre-defined parameters stored in
a second storage module of the controller;
e. allowing the graded object to pass through a duct (109), one by one, and
analyzing one or more variable factors by a plurality of sensors arranged
throughout the duct and (or) at least one line scan camera arranged at each
ejection point;
f. communicating to the controller, the one or more variable factors detected by
theat least one sensor (104)through a network controller and the one or more
variable factors detected by theat least one line scan camera (110) directlyfor
determining ejection points of each object;
g. actuating multiple ejectors (108) for ejecting objects by determining the
ejection points for the objects by analyzing the one or more variable factors
such as real-time position of each object, change in acceleration, velocity,
overtaking of a heavy/light object with respect to a hollow/light object and
behavior of the each object in the presence of turbulence; and
h. collecting the objects in multiple collecting locations of pre-defined grade from multiple outlets (106) configured opposite to the multiple ejectors (108). 9. The method for grading objects as claimed in claim 8, wherein the step of
communicating the one or more variable factors detected by the at least one sensor (104) to the controller comprises:

a. receiving signals from each sensor layer controllers of corresponding duct
related to the one or more variable factors of each object in real time to
determine the position, velocity and density of object accurately in real time in
said corresponding duct by the network controller of the sensor network of
corresponding duct (109);
b. sending said signals from said corresponding network controller of thesensor
network to the controller as each network controller of each duct is coupled to
the controller, and
c. receiving of the signals from the network controller of thesensor network of
corresponding duct by said controller and as the object cuts the multiple rays of
corresponding said sensor layers, thereby anticipating the exact position,
change in acceleration, velocity and density of each object accurately in real
time before the arrival of ejecting point of each conveying object during its
trajectory in corresponding duct by deciding ejecting point of each object.
10. The method for grading objects as claimed in claim 8, wherein the step of actuating multiple ejectors (108) for ejecting objects comprises:
a. receiving signals related to ejection of each object sent by the controller by the
ejectors (108) before the arrival of each grade of object in the corresponding
duct (109); and
b. jetting a beam of pre-defined duration of high pressure air or high pressure
fluid to eject the particular grade of object when a object of particular grade
comes near a particular grading point in the corresponding duct (109); and

c. making each grade of object to fall into the corresponding desired collecting location (107) through corresponding multiple outlets (106) for collecting different grades of objects into multiple grades in a single pass.