Photocurrent-based target object category identification method and system

Abstract

The invention discloses a photocurrent-based target object category identification method and system, and relates to the technical field of target classification and identification. The category identification method comprises the following steps: acquiring target infrared image data of a target object; determining a photocurrent data set of the target infrared image data according to the target infrared image data; determining a current data set according to the light current data set and a preset weight set corresponding to the light current data set; and determining target category data of the target object according to the current data set so as to determine the category of the target object according to the target category data. According to the category recognition system provided by the invention, the category of the target object is determined through the full-simulation neural network, and the use of a digital-to-analog converter can be avoided, so that the power consumption of the category recognition system can be effectively reduced.

Description

Technical Field

[0001] The present application relates to the technical field of target classification and recognition. Specifically, it relates to a method and system for classifying and recognizing a target object based on photocurrent. Background Art

[0002] Infrared detection technology, as an important part of the unmanned terminal detection system, shows application potential in the field of target classification and recognition. For example, infrared detection technology can be used for classifying and recognizing aircraft, cars, ships, etc.

[0003] Existing infrared detection systems rely on digital image processing technology. In this technical solution, the conversion process between analog signals and digital signals is involved, and analog-to-digital converter (ADC) devices are required, significantly increasing the power consumption of the recognition system.

[0004] The content of the background art part is only the technology known to the applicant and does not necessarily represent the prior art in this field. Summary of the Invention

[0005] The present application aims to provide a method and system for classifying and recognizing a target object based on photocurrent to solve the above technical problem of high system power consumption caused by using analog-to-digital conversion devices.

[0006] According to one aspect of the present application, there is provided a method for classifying and recognizing a target object based on photocurrent. The classification and recognition method includes: obtaining target infrared image data of the target object; determining a photocurrent data set of the target infrared image data according to the target infrared image data; determining a current data set according to the photocurrent data set and a preset weight set corresponding to the photocurrent data set; and determining target category data of the target object according to the current data set, so as to determine the category of the target object according to the target category data.

[0007] According to some embodiments of the present application, before the step of determining the current data set according to the photocurrent data set and the preset weight set corresponding to the photocurrent data set, the classification and recognition method further includes: determining the preset weight set.

[0008] According to some embodiments of the present application, the step of determining the current data set according to the photocurrent data set and the preset weight set corresponding to the photocurrent data set includes: performing amplification and conversion processing on the photocurrent data set to obtain a first voltage data set within a preset voltage range; determining current sub-data according to the first voltage data set and a preset weight subset corresponding to the first voltage data set; traversing all the preset weight subsets in the first voltage data set and the preset weight set to obtain all the current sub-data, so as to obtain the current data set according to all the current sub-data.

[0009] According to some embodiments of the present application, the step of determining the target category data of the target object based on the current data set includes: determining a second voltage data set based on the current data set; determining a differential voltage data set based on the second voltage data set; and determining the target category data based on the differential voltage data set.

[0010] According to some embodiments of the present application, the second voltage data set includes at least one second voltage sub-data set, and the second voltage sub-data set includes two second voltage sub-data; the step of determining the differential voltage data set based on the second voltage data set includes: determining a differential voltage sub-data based on the difference between the two second voltage sub-data of the second voltage sub-data set; traversing all the second voltage sub-data sets of the second voltage data set to obtain all the differential voltage sub-data, so as to obtain the differential voltage data set based on all the differential voltage sub-data.

[0011] According to some embodiments of the present application, the step of determining the target category data based on the differential voltage data set includes: comparing the differential voltage sub-data in the differential voltage data set; and determining the differential sub-voltage data that meets the preset category condition in the differential voltage data set as the target classification data.

[0012] According to one aspect of the present application, the present application provides a category recognition system for an object driven by photocurrent. The category recognition system includes an image acquisition module and a processing module. The image acquisition module acquires the target infrared image data of the target object, and determines the photocurrent data set of the target infrared image data according to the target infrared image data; the input end of the processing module is connected to the output end of the image acquisition module, determines the current data set according to the photocurrent data set and the preset weight set corresponding to the photocurrent data set, determines the target category data of the target object according to the current data set, so as to determine the category of the target object according to the target category data.

[0013] According to some embodiments of the present application, the category recognition system further includes a weight control module. The output end of the weight control module is connected to the input end of the processing module, determines the preset weight set, and outputs the preset weight set.

[0014] According to some embodiments of the present application, the processing module includes an amplification unit and a first processing unit. The input end of the amplification unit is connected to the output end of the image acquisition module, performs amplification and conversion processing on the photocurrent data set to obtain a first voltage data set within a preset voltage range; the input end of the first processing unit is connected to the output end of the amplification unit, determines the current sub-data according to the first voltage data set and the preset weight subset corresponding to the first voltage data set, traverses all the preset weight subsets in the first voltage data set and the preset weight set to obtain all the current sub-data, so as to obtain the current data set based on all the current sub-data.

[0015] According to some embodiments of the present application, the processing module further includes a current-voltage conversion unit, a differential circuit unit, and a second processing unit. The input end of the current-voltage conversion unit is connected to the output end of the first processing unit, and determines a second voltage data set according to the current data set; the input end of the differential circuit unit is connected to the output end of the current-voltage conversion unit, and determines a differential voltage data set according to the second voltage data; the input end of the second processing unit is connected to the output end of the differential circuit unit, and determines the target category data according to the differential voltage data set.

[0016] Beneficial effects The present application can determine a photocurrent data set of target infrared image data through the target infrared image data of the target object. The present application can determine a current data set through the photocurrent data set and a preset weight set corresponding to the photocurrent data set. The present application can determine the target category data of the target object through the current data set, so as to determine the category of the target object according to the target category data.

[0017] The category recognition system provided by the present application determines the category of the target object through a fully analog neural network, which can avoid using a digital-to-analog converter, thereby effectively reducing the power consumption of the category recognition system. Description of the drawings

[0018] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained according to these drawings.

[0019] Figure 1 Shows a schematic structural diagram of a category recognition system according to an embodiment of the present application; Figure 2 Shows another schematic structural diagram of a category recognition system according to an embodiment of the present application; Figure 3 Shows another schematic structural diagram of a category recognition system according to an embodiment of the present application; Figure 4 Shows a schematic flowchart of a category recognition method 1000 according to an embodiment of the present application; Figure 5 Shows another schematic flowchart of a category recognition method 1000 according to an embodiment of the present application; Figure 6 Shows a schematic flowchart of step S130 according to an embodiment of the present application; Figure 7 Shows a schematic flowchart of step S140 according to an embodiment of the present application; Figure 8Shows a schematic flowchart of step S142 according to an embodiment of the present application; Figure 9 Shows a schematic flowchart of step S143 according to an embodiment of the present application.

[0020] Description of reference numerals: Category recognition system 200.

[0021] Image acquisition module 210; processing module 220; weight control module 230.

[0022] Amplification unit 221; first processing unit 222; current-voltage conversion unit 223; differential circuit unit 224; second processing unit 225.

[0023] Near-infrared detector 211, Photocurrent amplifier 2211.

[0024] Memristor sub-unit 2221.

[0025] Current-voltage converter 2231.

[0026] Differential circuit 2241.

[0027] Weight writing module 231; weight switch group 232; Weight switch 2321. Detailed implementation manners

[0028] Example embodiments will now be described more fully with reference to the accompanying drawings. However, the example embodiments can be implemented in various forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided so that this application will be thorough and complete, and will fully convey the concept of the example embodiments to those skilled in the art. Like reference numerals in the figures denote the same or similar parts, and thus their repetitive description will be omitted.

[0029] The described features, structures or characteristics can be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to give a thorough understanding of the embodiments of the present disclosure. However, those skilled in the art will realize that the technical solutions of the present disclosure can be practiced without one or more of these specific details, or can be implemented in other ways, components, materials, devices, etc. In these cases, well-known structures, methods, devices, implementations, materials or operations will not be shown or described in detail.

[0030] In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that comprises a series of steps or units is not limited to the listed steps or units, but may optionally further include steps or units not listed, or may optionally further include other steps or units inherent to these processes, methods, products or devices.

[0031] In the description and claims of this application and the above drawings, the terms "first", "second", etc. are used to distinguish different objects, rather than to describe a specific order.

[0032] The technical solutions of this application will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of this application. Obviously, the described embodiments are part of the embodiments of this application, rather than all of them. All other embodiments obtained by those skilled in the art based on the embodiments in this application without creative efforts belong to the scope of protection of this application.

[0033] According to one aspect of this application, this application provides a category recognition system 200 for a target driven by photocurrent. Refer to Figure 1 , the category recognition system 200 may include an image acquisition module 210 and a processing module 220. Refer to Figure 2 , the category recognition system 200 may further include a weight control module 230. Refer to Figure 3 , the processing module 220 includes an amplification unit 221, a first processing unit 222, a current-voltage conversion unit 223, a differential circuit unit 224 and a second processing unit 225.

[0034] Below in conjunction with Figure 1 , Figure 2 and Figure 3 Describe a category recognition method 1000 for a target based on photocurrent provided by this application. The category recognition method 1000 may be executed by the category recognition system 200.

[0035] Refer to Figure 4 , the category recognition method 1000 may include step S110 - step S140.

[0036] In step S110, the image acquisition module 210 acquires target infrared image data of the target.

[0037] According to the exemplary embodiment, the target may be a target of the category to be recognized. The target infrared image data may be infrared image data of the target.

[0038] The image acquisition module 210 may acquire the target infrared image data through an infrared image sensor such as a near-infrared image sensor, a mid-infrared image sensor, a far-infrared image sensor, etc.

[0039] The type of the target object can be selected according to the recognition requirements, and the present application does not limit this here. For example, the target object can be three types of target objects to be recognized: an airplane, an oil barrel, and a ship.

[0040] The image acquisition module 210 is a near-infrared detector 211 in an i×j array. Here, i is the number of rows of the near-infrared detector 211 in the i×j array, and j is the number of columns of the near-infrared detector 211 in the i×j array. The target infrared image data acquired by the image acquisition module 210 is an infrared image in an i×j array. The number of pixels of the target infrared image data is m, and the value of m is the product of i and j.

[0041] For example, referring to Figure 3 , the image acquisition module 210 is a near-infrared detector 211 in an 8×8 array, and the image acquisition module 210 acquires the target infrared image data of an 8×8 array of the target object. The target infrared image data of the 8×8 array includes 64 pixel data.

[0042] In step S120, the image acquisition module 210 determines a photocurrent data set of the target infrared image data according to the target infrared image data.

[0043] According to the exemplary embodiment, one pixel of the target infrared image data is converted into one photocurrent sub-data. All the photocurrent sub-data form a photocurrent data set. The photocurrent data set can be a set of photocurrent sub-data converted according to the pixels of the target infrared image data.

[0044] The image acquisition module 210 can convert each pixel data in the target infrared image data into a corresponding photocurrent sub-data through a near-infrared image sensor, a mid-infrared image sensor, a far-infrared image sensor, etc., and form a photocurrent data set according to the photocurrent sub-data. The photocurrent data set can be a 1×m matrix.

[0045] Exemplarily, the image acquisition module 210 can convert 64 pixel data of the acquired target infrared image data of an 8×8 array into 64 photocurrent sub-data, and the 64 pixel data and the 64 photocurrent sub-data are in one-to-one correspondence. The image acquisition module 210 can form a photocurrent data set from the 64 photocurrent sub-data and output the photocurrent data set. The photocurrent data set can be a 1×64 matrix.

[0046] In step S130, the processing module 220 determines a current data set according to the photocurrent data set and a preset weight set corresponding to the photocurrent data set.

[0047] According to an exemplary embodiment, the preset weight set can be a set of preset weights corresponding to the photocurrent data set. The preset weight set can be an m×n matrix. Wherein, n is twice the number of target categories. The number of target categories is the number of types of target objects. The weights in the preset weight set correspond to the photocurrent sub-data.

[0048] For example, in the case where the target objects can be three types of objects to be recognized, namely an airplane, an oil barrel, and a ship, then n is 6.

[0049] The number of target categories of the target object can be set according to the recognition application scenario, and the present application does not make any restrictions. For example, the target object can also be any 5 or 10 types of objects, and then n corresponds to 10 or 20 respectively.

[0050] The current sub-data obtained by multiplying the photocurrent sub-data and the preset weights corresponding to the photocurrent sub-data, and all the current sub-data form a set of current sub-data, that is, the current data set. The current data set can be a set of current sub-data obtained by multiplying the photocurrent sub-data and the preset weights corresponding to the photocurrent sub-data. The current data set can be a 1×n matrix.

[0051] The input end of the processing module 220 is connected to the output end of the image acquisition module 210. The processing module 220 can determine the current data set according to the photocurrent data set and the preset weight set through a single-layer fully connected neural network.

[0052] For example, the processing module 220 can determine a current data set (1×6 matrix) according to a photocurrent data set (1×64 matrix) and a preset weight set (64×6 matrix) through a single-layer fully connected network.

[0053] In step S140, the processing module 220 determines the target category data of the target object according to the current data set, so as to determine the category of the target object according to the target category data.

[0054] According to an exemplary embodiment, the target category data can be the category type data of the target object. The processing module 220 can convert the current data set into a voltage data set, and perform a difference processing on two adjacent voltage data in the voltage data set, so as to obtain a differential voltage data set. The processing module 220 can compare each differential voltage sub-data in the differential voltage data set, and the differential voltage sub-data is the probability distribution of the predicted target category data.

[0055] The processing module 220 selects the maximum value in the differential voltage sub-data as the target category data, so as to determine the category of the target object according to the probability distribution of the target category data. The target category data represents the maximum probability of the predicted target object category.

[0056] Through the above embodiments, the present application can determine a photocurrent data set of target infrared image data based on the target infrared image data of the target object. The present application can determine a current data set based on the photocurrent data set and a preset weight set corresponding to the photocurrent data set. The present application can determine target category data of the target object based on the current data set, so as to determine the category of the target object according to the target category data.

[0057] The category recognition system provided by the present application determines the category of the target object through a fully analog neural network, which can avoid using a digital-to-analog converter, thereby effectively reducing the power consumption of the category recognition system.

[0058] Optionally, referring to Figure 5 , after step S120 and before step S130, the category recognition method 1000 may further include step S120a.

[0059] The weight control module 230 executes step S120a. Step S120a is a step of determining a preset weight set.

[0060] According to the exemplary embodiment, the output end of the weight control module 230 is connected to the input end of the processing module 220, and the weight control module 230 determines a preset weight set to output the preset weight set.

[0061] Referring to Figure 3 , the weight control module 230 may include a weight writing module 231 and a weight switch group 232.

[0062] According to the exemplary embodiment, the weight switch group 232 may include a plurality of weight switches 2321. The number of weight switches 2321 may be the sum of m and n. The number of row weight switches in the weight switch group 232 is m, and the number of column weight switches in the weight switch group 232 is n.

[0063] The weight writing module 231 can control the writing of the preset weight set by controlling the conduction and cutoff of the weight switch 2321. When the weight switch 2321 is selectively conducted, the weight writing module 231 writes the preset weight set into the processing module 220. When the weight switch 2321 is cutoff, the weight writing module 231 stops writing the preset weight set.

[0064] Optionally, referring to Figure 6 , step S130 may include steps S131 - S133.

[0065] In step S131, the amplification unit 221 performs amplification and conversion processing on the photocurrent data set to obtain a first voltage data set within a preset voltage range.

[0066] According to the exemplary embodiment, the input end of the amplification unit 221 is connected to the output end of the image acquisition module 210. The amplification unit 221 may be a 1×m array of photocurrent amplifiers 2211. The amplification unit 221 may be based on a transimpedance amplification circuit structure, and use capacitance compensation to stabilize the transimpedance amplification circuit. By comprehensively considering parameters such as gain, bandwidth, noise, stability, and power consumption, the circuit design with the lowest power consumption is selected as the target circuit design of the amplification unit 221.

[0067] Exemplarily, the transimpedance amplification circuit structure may be a forward amplification structure or a reverse amplification structure.

[0068] The first voltage sub-data obtained after the photocurrent sub-data is amplified and converted. All the first voltage sub-data form a first voltage data set. The first voltage data set may be a set of first voltage sub-data obtained after the photocurrent sub-data is amplified and converted. The first voltage data set includes at least one first voltage sub-data. The first voltage data set may be a 1×m matrix.

[0069] For example, the first voltage data set may be a 1×64 matrix.

[0070] The preset voltage range may be the voltage value range that the input end of the first processing unit 222 can withstand. For example, the preset voltage range may be the voltage value that the input end of the memristor analog computing circuit can withstand.

[0071] For example, the peak current signal of the photocurrent sub-data is about 100 nA, the voltage value that the input end of the memristor analog computing circuit can withstand is 0.1 V, and the gain of the amplification unit 221 using the transimpedance amplification circuit structure is 10 6 , according to the 8-bit infrared image noise-free test calculation method, the noise of the amplification unit 221 needs to be less than 390 μV.

[0072] In step S132, the first processing unit 222 determines the current sub-data according to the first voltage data set and the preset weight subset corresponding to the first voltage data set.

[0073] According to the exemplary embodiment, the input end of the first processing unit 222 is connected to the output end of the amplification unit 221. The first processing unit 222 may be an m×n array of memristor sub-units 2221, and the number of memristor sub-units 2221 may be the product of m and n.

[0074] The preset weight set may be a preset weight set corresponding to the first voltage data set. The preset weight set may be an m×n matrix.

[0075] The preset weight subset can be the weight subset corresponding to the current sub-data in the preset weight set. The preset weight subset can be an m×1 matrix. For example, the preset weight subset can be a 64×1 matrix.

[0076] The current sub-data can be the product of the first voltage data set and the preset weight subset, that is, the current sub-data can be the product of the first voltage data set (a 1×m matrix) and the preset weight subset (an m×1 matrix).

[0077] For example, the first processing unit 222 can determine the current sub-data according to the following formula: ; where, is a current sub-data; ( , …… ) is the first voltage data set. , , …… and are all the first voltage sub-data in the first voltage data set. ( , …… ) is a preset weight subset corresponding to the first voltage data set. is the preset weight value corresponding to in ( , ) corresponding to . is the preset weight value corresponding to in ( , ) corresponding to . is the preset weight value corresponding to in ( , ) corresponding to .

[0078] In step S133, the first processing unit 222 traverses all the preset weight subsets in the first voltage data set and the preset weight set to obtain all the current sub-data, so as to obtain the current data set according to all the current sub-data.

[0079] According to the exemplary embodiment, the first processing unit 222 can implement matrix multiply-accumulate operations according to Kirchhoff's current law. The first processing unit 222 can adopt a crossbar array structure, and the row data of the crossbar array structure (i.e., the input end of the first processing unit 222) is the first voltage data set. The column data of the crossbar array structure (i.e., the output end of the first processing unit 222) is the current data set. The crossbar array structure can implement ±1 binary weights by subtracting two columns of memristor weights (0 or 1).

[0080] The first processing unit 222 may perform a multiplication operation on the first voltage data set and all the preset weight subsets to obtain all the current sub-data. The first processing unit 222 obtains a current data set according to all the current sub-data.

[0081] For example, the first processing unit 222 may determine the current data set according to the following formula: ; where is another current sub-data; ( , ... ) is another preset weight subset corresponding to the first voltage data set.

[0082] is the preset weight value in ( , ... ) corresponding to . is the preset weight value in ( , ... ) corresponding to . is the preset weight value in ( , ... ) corresponding to .

[0083] The current data set may be a 1×n matrix, that is, ( , ... ). For example, the current data set may be a 1×6 matrix. The preset weight set may be: ; Each preset weight value in the preset weight set may be 0 or 1. Each preset weight value in the preset weight set may be trained according to the category data of the image of the pre-trained object. The category of the pre-trained object may be the same as the category of the target object.

[0084] The weight writing module 231 can be an MCU controller (Memory Controller Unit). The MCU controller can control the values of each preset weight value by selectively controlling the high or low level signals output by the MCU controller through the weight switch group 232. For example, when the MCU controller selects to control the weight switch 2321 in the first row and the first column to close and connect to the high level, the value of the preset weight value corresponding to the high level signal is 1. When the MCU controller selects to control the weight switch 2321 in the first row and the first column to close and connect to the low level, the value of the preset weight value corresponding to the low level signal is 0. The number of switch control signals of the MCU controller can be the sum of m and n. And the high and low level signals selectively controlled and output by the MCU controller correspond one-to-one to the preset weight values.

[0085] Through the above embodiments, the present application can perform amplification and conversion processing on the photocurrent data set to obtain the first voltage data set within a preset voltage range. The present application can determine the current sub-data through the first voltage data set and the preset weight subset corresponding to the first voltage data set. The present application can traverse all the preset weight subsets in the first voltage data set and the preset weight set to obtain all the current sub-data, thereby obtaining the current data set.

[0086] The category recognition system provided by the present application can use the photocurrent drive method and determine the preset weight set through binary weights, making the structure of the category recognition system of the present application simple and with low power consumption.

[0087] Optionally, referring to Figure 7 , step S140 may include steps S141 - S143.

[0088] In step S141, the current-voltage conversion unit 223 determines the second voltage data set according to the current data set.

[0089] According to the exemplary embodiment, the input end of the current-voltage conversion unit 223 is connected to the output end of the first processing unit 222. The current-voltage conversion unit 223 can be a 1×n array of current-voltage converters 2231.

[0090] For example, the current-voltage conversion unit 223 can be a 1×6 array of current-voltage converters 2231.

[0091] The second voltage data set can be a voltage set formed by converting the current sub-data into second voltage sub-data. The second voltage data set can be a 1×n matrix.

[0092] For example, the second voltage data set can be a 1×6 matrix.

[0093] The current-voltage conversion unit 223 converts all the current sub-data into corresponding second voltage sub-data, and all the second voltage sub-data form a second voltage data set.

[0094] In step S142, the differential circuit unit 224 determines a differential voltage data set according to the second voltage data set.

[0095] According to the exemplary embodiment, the input end of the differential circuit unit 224 is connected to the output end of the current-voltage conversion unit 223. The differential circuit unit 224 can be a 1×(n / 2) array of differential circuits 2241. For example, the differential circuit unit 224 can be a 1×3 array of differential circuits 2241.

[0096] The differential voltage sub-data is obtained by the difference between two adjacent second voltage sub-data. All the differential voltage sub-data form a differential voltage data set. The differential voltage data set can be a set of differential voltage sub-data obtained by the difference between two adjacent second voltage sub-data. The differential voltage data set can be a 1×(n / 2) matrix.

[0097] For example, the differential voltage data set can be a 1×3 matrix.

[0098] See Figure 3 , the two input ends of the differential circuit 2241 are respectively connected to the output ends of two current-voltage converters 2231, and these two current-voltage converters 2231 are adjacent.

[0099] The differential circuit 2241 subtracts two adjacent second voltage sub-data in the second voltage data to obtain differential voltage sub-data. All the differential voltage sub-data form a differential voltage data set.

[0100] Optionally, the second voltage data set includes at least one second voltage sub-data set, and the second voltage sub-data set includes two second voltage sub-data.

[0101] See Figure 8 , step S142 may include steps S1421 - S1422.

[0102] In step S1421, the differential circuit unit 224 determines differential voltage sub-data according to the difference between two second voltage sub-data in the second voltage sub-data set.

[0103] According to the exemplary embodiment, the second voltage sub-data set is a set formed by two adjacent second voltage sub-data in the second voltage data set in sequence. The second voltage sub-data set can be a 1×2 matrix. The two second voltage sub-data in the second voltage sub-data set are adjacent, and the serial number of the second second voltage sub-data is an even number.

[0104] For example, the first second voltage sub-data and the second second voltage sub-data in the second voltage data set form a second voltage sub-data set. The third second voltage sub-data and the fourth second voltage sub-data in the second voltage data set form a second voltage sub-data set. And so on, the (n - 1)th second voltage sub-data and the nth second voltage sub-data in the second voltage data set form a second voltage sub-data set.

[0105] The differential voltage sub-data can be the difference between two second voltage sub-data in a second voltage sub-data set.

[0106] One differential circuit 2241 in the differential circuit unit 224 can determine the differential voltage sub-data according to the difference between two second voltage sub-data in the second voltage sub-data set.

[0107] In step S1422, the differential circuit unit 224 traverses all the second voltage sub-data sets in the second voltage data set to obtain all the differential voltage sub-data, so as to obtain a differential voltage data set according to all the differential voltage sub-data.

[0108] According to the exemplary embodiment, the differential circuit 2241 can correspond to the second voltage sub-data set one by one.

[0109] Each differential circuit 2241 in the differential circuit unit 224 can determine the corresponding differential voltage sub-data according to the difference between two second voltage sub-data in the second voltage sub-data set. All the differential voltage sub-data form a differential voltage data set.

[0110] In step S143, the second processing unit 225 determines the target category data according to the differential voltage data set.

[0111] According to the exemplary embodiment, the input end of the second processing unit 225 is connected to the output end of the differential circuit unit 224. The second processing unit 225 can be a processor with data processing capabilities.

[0112] The input end of the second processing unit 225 can be connected to the output end of each differential circuit 2241. The second processing unit 225 can receive the differential voltage data set. The second processing unit 225 can compare all the differential voltage sub-data in the differential voltage data set and select the differential voltage sub-data with the maximum value as the target category data. The maximum value can be the positive maximum value or the negative maximum value.

[0113] Optionally, referring to Figure 9 , step S143 can include steps S1431 - S1432.

[0114] In step S1431, the second processing unit 225 compares the differential voltage sub-data in the differential voltage data set.

[0115] According to the exemplary embodiment, the value of the differential voltage sub-data can be positive, negative, or zero. Each differential voltage sub-data corresponds to the probability of the predicted target category data.

[0116] In the case where the transimpedance amplifier circuit structure is a forward amplification structure, the second processing unit 225 can directly compare the differential voltage sub-data.

[0117] In the case where the transimpedance amplifier circuit structure is a reverse amplification structure, the second processing unit 225 can add a negative sign before the differential voltage sub-data and then make a comparison.

[0118] In step S1432, the second processing unit 225 determines the differential sub-voltage data that meets the preset category condition in the differential voltage data set as the target classification data.

[0119] According to the exemplary embodiment, the preset category condition rule can be the maximum positive value of the differential voltage sub-data. The preset category condition rule can also be the maximum negative value of the differential voltage sub-data.

[0120] In the case where the transimpedance amplifier circuit structure is a forward amplification structure, the preset category condition rule can be the maximum positive value of the differential voltage sub-data.

[0121] For example, in the case where the transimpedance amplifier circuit structure can be a forward amplification structure, the values of the three differential voltage sub-data in the differential voltage data set are -1, 1, and -3 respectively. Then, the second processing unit 225 determines that the category corresponding to the differential voltage sub-data with a value of 1 is the target category data.

[0122] In the case where the transimpedance amplifier circuit structure is a reverse amplification structure, the preset category condition rule can be the maximum negative value of the differential voltage sub-data.

[0123] For example, in the case where the transimpedance amplifier circuit structure can be a reverse amplification structure, the values of the three differential voltage sub-data in the differential voltage data set are -1, 1, and -3 respectively. Then, the second processing unit 225 determines that the category corresponding to the differential voltage sub-data with a value of -3 is the target category data.

[0124] Through the above embodiments, the present application can determine the second voltage data set through the current data set, the present application can determine the differential voltage data set through the second voltage data set, and the present application can determine the target category data through the differential voltage data set.

[0125] The processing module circuit structure of the category recognition system provided by this application is simple, and the processing module operates by photocurrent drive, which can avoid the circuit that converts the photocurrent in the image acquisition module into other digital signal forms, thereby further reducing the power consumption of the category recognition system.

[0126] According to the exemplary embodiment, the number of target categories of the category recognition system provided by this application can be downward compatible. For example, if the preset number of target categories of the category recognition system is 10, then the category recognition system can recognize any number of target objects from 1 to 10.

[0127] Finally, it should be noted that the above are only the preferred embodiments of this application and are not intended to limit this application. Although this application has been described in detail with reference to the foregoing embodiments, for those skilled in the art, they can still modify the technical solutions of the foregoing embodiments, or perform equivalent replacements for some of the technical features. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of this application shall be included within the protection scope of this application.

Claims

1. A method for classifying a target based on photocurrent, characterized in that: The category identification method comprises: Acquiring target infrared image data of the target object; Determining a photocurrent data set of the target infrared image data according to the target infrared image data; Determining a current data set according to the photocurrent data set and a preset weight set corresponding to the photocurrent data set; The target category data of the target object is determined according to the current data set, so as to determine the category of the target object according to the target category data.

2. The category identification method according to claim 1, characterized in that: Before determining the current data set according to the photocurrent data set and the preset weight set corresponding to the photocurrent data set, the category identification method further includes: The preset weight set is determined.

3. The category identification method according to claim 1, characterized in that: The determining of the current data set according to the photocurrent data set and a preset weight set corresponding to the photocurrent data set comprises: amplifying and converting the photocurrent data set to obtain a first voltage data set within a preset voltage range; Determining current sub-data according to the first voltage data set and a preset weight subset corresponding to the first voltage data set; The first voltage data set and all preset weight subsets in the preset weight set are traversed to obtain all current sub-data, so as to obtain the current data set according to all the current sub-data.

4. The category identification method according to claim 1, characterized in that: Determining the target category data of the target object according to the current data set includes: determining a second voltage data set based on the current data set; determining a differential voltage data set based on the second voltage data set; The target category data is determined according to the differential voltage data set.

5. The category identification method according to claim 4, characterized in that: The second voltage data set includes at least one second voltage sub-data set, and the second voltage sub-data set includes two second voltage sub-data; Determining a differential voltage data set according to the second voltage data set includes: determining differential voltage sub-data according to a difference between the two second voltage sub-data of the second voltage sub-data set; All second voltage sub-data sets of the second voltage data set are traversed to obtain all differential voltage sub-data, so as to obtain the differential voltage data set according to all the differential voltage sub-data.

6. The category identification method according to claim 4, characterized in that: The determining the target category data according to the differential voltage data set comprises: comparing the differential voltage sub-data in the differential voltage data set; The differential sub-voltage data satisfying a preset classification condition in the differential voltage data sets are determined as the target classification data.

7. A photocurrent driven target classification system, characterized in that: The category identification system comprises: An image acquisition module is used to acquire target infrared image data of the target object, and determine a photocurrent data set of the target infrared image data according to the target infrared image data; A processing module, whose input end is connected to the output end of the image acquisition module, determines a current data set according to the photocurrent data set and a preset weight set corresponding to the photocurrent data set, determines target category data of the target object according to the current data set, and determines the category of the target object according to the target category data.

8. The category identification system according to claim 7, characterized in that: The category identification system further includes: A weight control module, whose output end is connected to the input end of the processing module, determines the preset weight set and outputs the preset weight set.

9. The category identification system according to claim 7, characterized in that: The processing module comprises: an amplification unit, whose input end is connected to the output end of the image acquisition module, and performs amplification and conversion processing on the photocurrent data set to obtain a first voltage data set within a preset voltage range; A first processing unit, whose input end is connected to the output end of the amplifying unit, determines the current sub-data according to the first voltage data set and the preset weight subset corresponding to the first voltage data set, traverses the first voltage data set and all the preset weight subsets in the preset weight set to obtain all the current sub-data, so as to obtain the current data set according to all the current sub-data.

10. The category identification system according to claim 9, characterized in that: The processing module also includes: a current-to-voltage conversion unit, whose input end is connected to the output end of the first processing unit, and determines a second voltage data set according to the current data set; a differential circuit unit, the input end of which is connected to the output end of the current-voltage conversion unit, and determines a differential voltage data set according to the second voltage data; The second processing unit has an input end connected to the output end of the differential circuit unit and determines the target category data according to the differential voltage data set.

Images