Target object identification method, device, and equipment, and storage medium

Abstract

The application provides a target object identification method and device, equipment and a storage medium, and relates to the technical field of image processing. The method comprises the following steps: acquiring a frame image of a target object to be identified; determining a first feature matrix and a second feature matrix in the frame image of the target object to be identified; determining a third feature matrix and a fourth feature matrix according to the first feature matrix and the second feature matrix; wherein the third feature matrix represents an updated first feature matrix; the fourth feature matrix represents an updated second feature matrix; performing feature fusion on the third feature matrix and the fourth feature matrix to determine a feature fusion result; and identifying the target object to be identified in the frame image of the target object to be identified according to the feature fusion result. The technical scheme can improve the accuracy of small target object detection and avoid missed detection.

Description

Technical Field

[0001] This application relates to the field of image processing technology, and in particular to a method, apparatus, device and storage medium for target object recognition. Background Technology

[0002] Currently, in bank security monitoring scenarios, there are numerous tasks involving the detection of small target samples, including warnings for unattended cash at teller counters and fire prevention alerts. Small target samples refer to objects whose area of ​​measurement is less than 10% of the input image size. The challenge lies in the fact that the small size of the target makes it easy to miss detections, or the low resolution can lead to misjudgments.

[0003] Therefore, there is an urgent need for a target object recognition method that can improve the accuracy of small target sample detection and thus avoid missed detections. Summary of the Invention

[0004] This application provides a target object identification method, apparatus, device, and storage medium, which can improve the accuracy of small target sample detection and thus avoid missed detections.

[0005] In a first aspect, this application provides a method for identifying a target object, including:

[0006] A frame image of the target object to be identified is acquired, and a first feature matrix and a second feature matrix are determined in the frame image of the target object to be identified; wherein, the first feature matrix represents semantic information; and the second feature matrix represents resolution information.

[0007] Based on the first feature matrix and the second feature matrix, a third feature matrix and a fourth feature matrix are determined; wherein, the third feature matrix represents the updated first feature matrix; and the fourth feature matrix represents the updated second feature matrix.

[0008] The third feature matrix and the fourth feature matrix are fused to determine the result of the feature fusion.

[0009] Based on the result of the feature fusion, the target object to be identified is identified in the frame image of the target object to be identified.

[0010] In one example, determining the third and fourth feature matrices based on the first and second feature matrices includes:

[0011] The mask value is determined based on the first feature matrix and the second feature matrix;

[0012] Based on the mask value, the first feature matrix and the second feature matrix are processed to obtain the third feature matrix and the fourth feature matrix.

[0013] In one example, processing the first feature matrix and the second feature matrix according to the mask value to obtain the third feature matrix and the fourth feature matrix includes:

[0014] Based on the mask value, a first scale transformation is performed on the first feature matrix and the second feature matrix to determine the third feature matrix;

[0015] Based on the mask value, a second scale transformation is performed on the first feature matrix and the second feature matrix to determine the fourth feature matrix.

[0016] In one example, determining the third feature matrix by performing a first scale transformation on the first feature matrix and the second feature matrix based on the mask value includes:

[0017] The mask values ​​are transformed by spatial mapping to obtain new mask values;

[0018] Based on the new mask value, the first parameter value, and the second parameter value, the first feature matrix and the second feature matrix are subjected to a first scale transformation to obtain the first result value after the first scale transformation.

[0019] The third feature matrix is ​​determined based on the first result value.

[0020] In one example, the step of performing a second-scale transformation on the first feature matrix and the second feature matrix based on the mask value to determine the fourth feature matrix includes:

[0021] The mask values ​​are transformed by spatial mapping to obtain new mask values;

[0022] Based on the new mask value, the first parameter value, and the third parameter value, a second scale transformation is performed on the first feature matrix and the second feature matrix to obtain the second result value after the second scale transformation.

[0023] The fourth feature matrix is ​​determined based on the second result value.

[0024] In one example, the feature fusion of the third feature matrix and the fourth feature matrix, and the determination of the feature fusion result, includes:

[0025] Adjust the first weight information of the third feature matrix and the second weight information of the fourth feature matrix; wherein the sum of the first weight information and the second weight information is 1;

[0026] The result of feature fusion is determined based on the third feature matrix, the first weight information, the fourth feature matrix, and the second weight information.

[0027] In one example, identifying the target object in the frame image of the target object to be identified based on the result of the feature fusion includes:

[0028] Based on the result of the feature fusion, the target object to be identified is labeled in the frame image of the target object to be identified.

[0029] Secondly, this application provides a target object recognition device, the device comprising:

[0030] The first determining unit is configured to acquire a frame image of a target object to be identified, and determine a first feature matrix and a second feature matrix in the frame image of the target object to be identified; wherein, the first feature matrix represents semantic information; and the second feature matrix represents resolution information.

[0031] The second determining unit is configured to determine a third feature matrix and a fourth feature matrix based on the first feature matrix and the second feature matrix; wherein the third feature matrix represents the updated first feature matrix; and the fourth feature matrix represents the updated second feature matrix.

[0032] The third determining unit is used to perform feature fusion on the third feature matrix and the fourth feature matrix, and determine the result of feature fusion.

[0033] The recognition unit is used to identify the target object in the frame image of the target object to be identified based on the result of the feature fusion.

[0034] In one example, the second determining unit includes:

[0035] The first determining module is used to determine the mask value based on the first feature matrix and the second feature matrix;

[0036] The second determining module is used to process the first feature matrix and the second feature matrix according to the mask value to obtain the third feature matrix and the fourth feature matrix.

[0037] In one example, the second determining module includes:

[0038] The first determining submodule is used to perform a first scale transformation on the first feature matrix and the second feature matrix according to the mask value to determine the third feature matrix;

[0039] The second determining submodule is used to perform a second scale transformation on the first feature matrix and the second feature matrix based on the mask value to determine the fourth feature matrix.

[0040] In one example, the first determining submodule is used to perform a spatial mapping transformation on the mask value to obtain a new mask value; based on the new mask value, the first parameter value, and the second parameter value, perform a first scale transformation on the first feature matrix and the second feature matrix to obtain a first result value after the first scale transformation; and determine the third feature matrix based on the first result value.

[0041] In one example, the second determining submodule is specifically used to perform a spatial mapping transformation on the mask value to obtain a new mask value; based on the new mask value, the first parameter value, and the third parameter value, perform a second scale transformation on the first feature matrix and the second feature matrix to obtain a second result value after the second scale transformation; and determine the fourth feature matrix based on the second result value.

[0042] In one example, the third determining unit includes:

[0043] An adjustment module is used to adjust the first weight information of the third feature matrix and the second weight information of the fourth feature matrix; wherein the sum of the first weight information and the second weight information is 1.

[0044] The determining module is used to determine the result of feature fusion based on the third feature matrix, the first weight information, the fourth feature matrix, and the second weight information.

[0045] In one example, the identification unit is specifically used to label the target object to be identified in the frame image of the target object to be identified based on the result of the feature fusion.

[0046] Thirdly, this application provides an electronic device, including: a processor, and a memory communicatively connected to the processor;

[0047] The memory stores computer-executed instructions;

[0048] The processor executes computer execution instructions stored in the memory to implement the method as described in the first aspect.

[0049] Fourthly, this application provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the method described in the first aspect.

[0050] Fifthly, this application provides a computer program product, including a computer program that, when executed by a processor, implements the method described in the first aspect.

[0051] This application provides a target object recognition method, apparatus, device, and storage medium. The method involves acquiring a frame image of a target object to be identified, determining a first feature matrix and a second feature matrix in the frame image, wherein the first feature matrix represents semantic information and the second feature matrix represents resolution information; determining a third feature matrix and a fourth feature matrix based on the first and second feature matrices, wherein the third feature matrix represents the updated first feature matrix and the fourth feature matrix represents the updated second feature matrix; performing feature fusion on the third and fourth feature matrices to determine the result of feature fusion; and identifying the target object in the frame image based on the result of feature fusion. This technical solution improves the accuracy of small target sample detection, thereby avoiding missed detections. Attached Figure Description

[0052] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0053] Figure 1 This is a flowchart illustrating a target object identification method according to Embodiment 1 of this application;

[0054] Figure 2 This is a flowchart illustrating a target object identification method according to Embodiment 2 of this application;

[0055] Figure 3 This is a schematic diagram of the structure of a target object recognition device according to Embodiment 3 of this application;

[0056] Figure 4 This is a schematic diagram of the structure of a target object recognition device according to Embodiment 4 of this application;

[0057] Figure 5 This is a block diagram illustrating an electronic device according to an exemplary embodiment.

[0058] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0059] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0060] The target object identification method provided in this application aims to solve the above-mentioned technical problems in the prior art.

[0061] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0062] Figure 1 This is a flowchart illustrating a target object identification method according to Embodiment 1 of this application. Embodiment 1 includes the following steps:

[0063] S101. Obtain the frame image of the target object to be identified, and determine the first feature matrix and the second feature matrix in the frame image of the target object to be identified; wherein, the first feature matrix represents semantic information; and the second feature matrix represents resolution information.

[0064] In one example, the target object to be identified is an object that occupies a relatively small proportion of the frame image, such as a hundred-yuan bill on a counter or a flame.

[0065] In this embodiment, the frame image of the target object to be identified is input into a preset model. Feature maps of the frame image of the target object to be identified can be extracted, and a first feature matrix and a second feature matrix can be determined based on the feature maps. The first feature matrix is ​​extracted from the higher layers of the neural network. Specifically, the first feature matrix represents semantic information and can be the specific physical form of the target object to be identified. For example, the target object to be identified is a banknote. The second feature matrix is ​​extracted from the lower layers of the neural network. Specifically, the second feature matrix represents resolution information and can be the edge features, color features, and initial position features of the target object to be identified.

[0066] S102. Based on the first and second characteristic matrices, determine the third and fourth characteristic matrices; wherein the third characteristic matrix represents the updated first characteristic matrix; and the fourth characteristic matrix represents the updated second characteristic matrix.

[0067] In this embodiment, the third feature matrix is ​​obtained by processing the first feature matrix, and the fourth feature matrix is ​​obtained by processing the second feature matrix. The third feature matrix represents semantic information, specifically the physical form of the target object to be identified. The fourth feature matrix represents resolution information, which may include the edge features, color features, and initial position features of the target object.

[0068] S103. Perform feature fusion on the third and fourth feature matrices to determine the result of feature fusion.

[0069] In this embodiment, the third and fourth feature matrices are feature descriptions of the target object to be identified at different scales. Therefore, the third and fourth feature matrices can be fused to obtain a more detailed feature description. Specific feature fusion algorithms can include positional addition and merging, resulting in a fused feature description.

[0070] S104. Based on the result of feature fusion, identify the target object in the frame image of the target object to be identified.

[0071] In this embodiment, after obtaining the feature fusion result, the target object to be identified is labeled in the frame image of the target object. Specifically, the labeling method can be to label with rectangular boxes.

[0072] This application provides a target object recognition method. The method involves acquiring a frame image of the target object to be identified, determining a first feature matrix and a second feature matrix within the frame image, determining a third feature matrix and a fourth feature matrix based on the first and second feature matrices, fusing the third and fourth feature matrices to determine the result of the feature fusion, and identifying the target object in the frame image based on the result of the feature fusion. This technical solution allows for the selection and fusion of features of different sizes and can be applied to any multi-scale feature fusion network and its related variants. It also exhibits high reusability in security scenarios. Furthermore, this method is data-independent and does not alter the end-to-end neural network training method, maintaining the computational efficiency of the security detection process while improving the detection rate of small target objects.

[0073] Figure 2 This is a flowchart illustrating a target object identification method according to Embodiment 2 of this application. Embodiment 1 includes the following steps:

[0074] S201. Obtain a frame image of the target object to be identified, and determine the first feature matrix and the second feature matrix in the frame image of the target object to be identified; wherein, the first feature matrix represents semantic information; and the second feature matrix represents resolution information.

[0075] For example, this step can refer to step S101 above, and will not be repeated here.

[0076] S202. Determine the mask value based on the first feature matrix and the second feature matrix.

[0077] In this embodiment, the first feature matrix and the second feature matrix are merged through channels to obtain the fused scale features, and the spatial average pooling method is used to obtain the mask value.

[0078] S203. Based on the mask value, process the first feature matrix and the second feature matrix to obtain the third feature matrix and the fourth feature matrix.

[0079] In this embodiment, the mask value is adaptively learned based on the mask value, and the information distribution of different channels is readjusted.

[0080] In one example, based on the mask values, the first and second feature matrices are processed to obtain the third and fourth feature matrices, including:

[0081] Based on the mask values, a first scale transformation is performed on the first and second feature matrices to determine the third feature matrix; based on the mask values, a second scale transformation is performed on the first and second feature matrices to determine the fourth feature matrix.

[0082] In this embodiment, the third feature matrix is ​​obtained by performing a first scale transformation on the mask value, the first feature matrix, and the second feature matrix. The first scale transformation can be performed by bitwise multiplication.

[0083] After performing a second scale transformation on the mask value, the first feature matrix, and the second feature matrix, the fourth feature matrix is ​​obtained. The second scale transformation can be performed by bitwise multiplication.

[0084] In this embodiment, the third feature matrix is ​​jointly determined by the first and second feature matrices; therefore, the information represented by the third feature matrix is ​​more accurate. Each set of mask values ​​is multiplied bitwise with its corresponding feature matrix to obtain a new adaptive feature distribution. That is, compared to the initial feature matrices, the two sets of feature matrices further readjust their information distribution based on each other's information.

[0085] In one example, based on the mask values, a first-scale transformation is performed on the first and second feature matrices to determine the third feature matrix, including:

[0086] The mask values ​​are transformed by spatial mapping to obtain new mask values; based on the new mask values, the first parameter value, and the second parameter value, the first feature matrix and the second feature matrix are transformed by the first scale to obtain the first result value after the first scale transformation; based on the first result value, the third feature matrix is ​​determined.

[0087] Specifically, the third characteristic matrix can be obtained using the following formula:

[0088] h'=σ(φ([h; l], θ1), θ2)·h;

[0089] Where h represents the first feature matrix and l represents the second feature matrix. θ1 represents the first parameter value, θ2 represents the second parameter value, and σ represents the sigmoid activation function. h' represents the activation function ReLU, and h' represents the third feature matrix.

[0090] In one example, based on the mask values, a second scaling transformation is performed on the first and second feature matrices to determine the fourth feature matrix, including:

[0091] The mask values ​​are transformed by spatial mapping to obtain new mask values;

[0092] Based on the new mask value, the first parameter value, and the third parameter value, a second scale transformation is performed on the first feature matrix and the second feature matrix to obtain the second result value after the second scale transformation.

[0093] The fourth feature matrix is ​​determined based on the second result value.

[0094] Specifically, the fourth characteristic matrix can be obtained using the following formula:

[0095] l'=σ(φ([h;l],θ1),θ3)l;

[0096] In this embodiment, h represents the first feature matrix, and l represents the second feature matrix. Here, θ1 represents the first parameter value, θ3 represents the third parameter value, and σ represents the sigmoid activation function. l represents the activation function ReLU, and l' represents the fourth feature matrix.

[0097] S204. Adjust the first weight information of the third feature matrix and the second weight information of the fourth feature matrix; wherein the sum of the first weight information and the second weight information is 1.

[0098] In this embodiment, the larger the value of the second weight information, the larger the proportion of the fourth feature matrix, and the more accurate the information obtained when the target object to be identified is a relatively small object. In this embodiment, after average pooling of the fourth feature matrix, the mask value after average pooling is determined. Then, a nonlinear transformation is performed on the mask value after average pooling, and then calculation is performed through a fully connected layer. First, the mask value after average pooling is compressed to a low-dimensional space, and then stretched to a high-dimensional space to determine the high-dimensional mask value. After a nonlinear transformation is performed on the high-dimensional mask value, the second weight information is obtained. The first weight information is the difference between the value 1 and the second weight information.

[0099] S205. Determine the result of feature fusion based on the third feature matrix, the first weight information, the fourth feature matrix, and the second weight information.

[0100] In this embodiment, the third feature matrix is ​​multiplied digit-wise by the first weight information, and the fourth feature matrix is ​​multiplied digit-wise by the second weight information. The results are then added digit-wise. For details, please refer to the following formula:

[0101] t = σ(l', θ4);

[0102] F = (1-t)·h' + t·l';

[0103] Where t represents the second weight information, 1-t represents the first weight information, θ4 represents the fourth parameter value, h' represents the third feature matrix, and l' represents the fourth feature matrix.

[0104] S206. Based on the result of feature fusion, label the target object to be identified in the frame image of the target object to be identified.

[0105] For example, this step can refer to step S104 above, and will not be repeated here.

[0106] This application provides a target object recognition method that can readjust the first weight information of the third feature matrix and the second weight information of the fourth feature matrix. It can control the redistribution of high-order features through low-order features, thereby increasing the information proportion of the current small-scale target object. Secondly, through optional feature fusion, it can further select the proportion within the scale range of the current small target object, thereby weakening the side effect information, strengthening the quantitative expression of small target features, and improving the detection effect of small target objects.

[0107] Figure 3 This is a schematic diagram of a target object recognition device according to Embodiment 3 of this application. Specifically, the device 30 in Embodiment 3 includes:

[0108] The first determining unit 301 is used to acquire a frame image of the target object to be identified and to determine a first feature matrix and a second feature matrix in the frame image of the target object to be identified; wherein, the first feature matrix represents semantic information and the second feature matrix represents resolution information.

[0109] The second determining unit 302 is used to determine a third feature matrix and a fourth feature matrix based on the first feature matrix and the second feature matrix; wherein the third feature matrix represents the updated first feature matrix; and the fourth feature matrix represents the updated second feature matrix.

[0110] The third determining unit 303 is used to perform feature fusion on the third feature matrix and the fourth feature matrix to determine the result of feature fusion.

[0111] The recognition unit 304 is used to identify the target object in the frame image of the target object to be identified based on the result of feature fusion.

[0112] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the above-described device can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0113] Figure 4 This is a schematic diagram of a target object recognition device according to Embodiment 4 of this application. Specifically, the device 40 in Embodiment 4 includes:

[0114] The first determining unit 401 is used to acquire a frame image of the target object to be identified and to determine a first feature matrix and a second feature matrix in the frame image of the target object to be identified; wherein, the first feature matrix represents semantic information and the second feature matrix represents resolution information.

[0115] The second determining unit 402 is used to determine a third feature matrix and a fourth feature matrix based on the first feature matrix and the second feature matrix; wherein the third feature matrix represents the updated first feature matrix; and the fourth feature matrix represents the updated second feature matrix.

[0116] The third determining unit 403 is used to perform feature fusion on the third feature matrix and the fourth feature matrix to determine the result of feature fusion.

[0117] The recognition unit 404 is used to identify the target object in the frame image of the target object to be identified based on the result of feature fusion.

[0118] In one example, the second determining unit 402 includes:

[0119] The first determining module 4021 is used to determine the mask value based on the first feature matrix and the second feature matrix;

[0120] The second determining module 4022 is used to process the first feature matrix and the second feature matrix according to the mask value to obtain the third feature matrix and the fourth feature matrix.

[0121] In one example, the second determining module 4022 includes:

[0122] The first determining submodule 40221 is used to perform a first scale transformation on the first feature matrix and the second feature matrix according to the mask value to determine the third feature matrix;

[0123] The second determining submodule 40222 is used to perform a second scale transformation on the first feature matrix and the second feature matrix based on the mask value to determine the fourth feature matrix.

[0124] In one example, the first determining submodule 40221 is used to perform a spatial mapping transformation on the mask value to obtain a new mask value; based on the new mask value, the first parameter value, and the second parameter value, a first scale transformation is performed on the first feature matrix and the second feature matrix to obtain a first result value after the first scale transformation; and based on the first result value, a third feature matrix is ​​determined.

[0125] In one example, the second determining submodule 40222 is specifically used to perform a spatial mapping transformation on the mask value to obtain a new mask value; based on the new mask value, the first parameter value, and the third parameter value, perform a second scale transformation on the first feature matrix and the second feature matrix to obtain a second result value after the second scale transformation; and determine the fourth feature matrix based on the second result value.

[0126] In one example, the third determining unit 403 includes:

[0127] The adjustment module 4031 is used to adjust the first weight information of the third feature matrix and the second weight information of the fourth feature matrix; wherein the sum of the first weight information and the second weight information is 1.

[0128] The determination module 4032 is used to determine the result of feature fusion based on the third feature matrix, the first weight information, the fourth feature matrix, and the second weight information.

[0129] In one example, the recognition unit 404 is specifically used to annotate the target object to be recognized in the frame image of the target object to be recognized based on the result of feature fusion.

[0130] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the above-described device can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0131] Figure 5This is a block diagram illustrating an electronic device according to an exemplary embodiment. The device may be a mobile phone, computer, digital broadcasting terminal, messaging device, game console, tablet device, medical device, fitness device, personal digital assistant, etc.

[0132] The device 500 may include one or more of the following components: a processing component 502, a memory 504, a power supply component 506, a multimedia component 508, an audio component 510, an input / output (I / O) interface 512, a sensor component 514, and a communication component 516.

[0133] Processing component 502 typically controls the overall operation of device 500, such as operations associated with display, telephone calls, data communication, camera operation, and recording. Processing component 502 may include one or more processors 520 to execute instructions to complete all or part of the steps of the methods described above. Furthermore, processing component 502 may include one or more modules to facilitate interaction between processing component 502 and other components. For example, processing component 502 may include a multimedia module to facilitate interaction between multimedia component 508 and processing component 502.

[0134] Memory 504 is configured to store various types of data to support the operation of device 500. Examples of such data include instructions for any application or method operating on device 500, contact data, phonebook data, messages, pictures, videos, etc. Memory 504 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0135] Power supply component 506 provides power to various components of device 500. Power supply component 506 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power to device 500.

[0136] Multimedia component 508 includes a screen that provides an output interface between the device 500 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touchscreen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touches, swipes, and gestures on the touch panel. The touch sensors may sense not only the boundaries of the touch or swipe action but also the duration and pressure associated with the touch or swipe operation. In some embodiments, multimedia component 508 includes a front-facing camera and / or a rear-facing camera. When the device 500 is in an operating mode, such as a shooting mode or a video mode, the front-facing camera and / or the rear-facing camera may receive external multimedia data. Each front-facing camera and rear-facing camera may be a fixed optical lens system or have focal length and optical zoom capabilities.

[0137] Audio component 510 is configured to output and / or input audio signals. For example, audio component 510 includes a microphone (MIC) configured to receive external audio signals when device 500 is in an operating mode, such as call mode, recording mode, and voice recognition mode. The received audio signals may be further stored in memory 504 or transmitted via communication component 516. In some embodiments, audio component 510 also includes a speaker for outputting audio signals.

[0138] I / O interface 512 provides an interface between processing component 502 and peripheral interface modules, such as keyboards, click wheels, buttons, etc. These buttons may include, but are not limited to, home buttons, volume buttons, power buttons, and lock buttons.

[0139] Sensor assembly 514 includes one or more sensors for providing status assessments of various aspects of device 500. For example, sensor assembly 514 may detect the on / off state of device 500, the relative positioning of components such as the display and keypad of device 500, changes in the position of device 500 or a component of device 500, the presence or absence of user contact with device 500, the orientation or acceleration / deceleration of device 500, and temperature changes of device 500. Sensor assembly 514 may include a proximity sensor configured to detect the presence of nearby objects without any physical contact. Sensor assembly 514 may also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, sensor assembly 514 may also include an accelerometer, a gyroscope, a magnetometer, a pressure sensor, or a temperature sensor.

[0140] Communication component 516 is configured to facilitate wired or wireless communication between device 500 and other devices. Device 500 can access wireless networks based on communication standards, such as WiFi, 2G, or 3G, or combinations thereof. In one exemplary embodiment, communication component 516 receives broadcast signals or broadcast-related information from an external broadcast management system via a broadcast channel. In one exemplary embodiment, communication component 516 also includes a near-field communication (NFC) module to facilitate short-range communication. For example, the NFC module may be implemented based on radio frequency identification (RFID) technology, Infrared Data Association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.

[0141] In an exemplary embodiment, the apparatus 500 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the methods described above.

[0142] In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions is also provided, such as a memory 504 including instructions, which can be executed by a processor 520 of the device 500 to perform the above-described method. For example, the non-transitory computer-readable storage medium may be a ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device, etc.

[0143] A non-transitory computer-readable storage medium, wherein instructions in the storage medium, when executed by a processor of an electronic device, enable the electronic device to perform the target object identification method described above.

[0144] This application also discloses a computer program product, including a computer program that, when executed by a processor, implements the method described in this embodiment.

[0145] Various embodiments of the systems and technologies described above in this application can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include: implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0146] The program code used to implement the methods of this application may be written in any combination of one or more programming languages. This program code may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing device, such that when executed by the processor or controller, the functions / operations specified in the flowcharts and / or block diagrams are implemented. The program code may be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or electronic device.

[0147] In the context of this application, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. Machine-readable media can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0148] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0149] The systems and technologies described herein can be implemented in computing systems that include back-end components (e.g., as data electronic devices), or computing systems that include middleware components (e.g., application electronic devices), or computing systems that include front-end components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with implementations of the systems and technologies described herein), or any combination of such back-end, middleware, or front-end components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., a communication network). Examples of communication networks include local area networks (LANs), wide area networks (WANs), and the Internet.

[0150] Computer systems can include client and electronic devices. Clients and electronic devices are generally geographically separated and typically interact via communication networks. The client-electronic device relationship is created by computer programs running on the respective computers and having a client-electronic device relationship with each other. The electronic device can be a cloud electronic device, also known as a cloud computing electronic device or cloud host, a host product within the cloud computing service system, addressing the shortcomings of traditional physical hosts and VPS services ("Virtual Private Server," or simply "VPS") in terms of management difficulty and weak business scalability. The electronic device can also be an electronic device in a distributed system or an electronic device incorporating blockchain technology. It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this application can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this application is achieved, and this is not limited herein.

[0151] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this application are indicated by the following claims.

[0152] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this application is limited only by the appended claims.

Claims

1. A target object recognition method characterized by, The method comprises: acquiring a frame image of a target object to be identified, determining a first feature matrix and a second feature matrix in the frame image of the target object to be identified; wherein the first feature matrix represents semantic information; the second feature matrix represents resolution information, and the target object to be identified is an object in the frame image whose size ratio is less than a preset value; determining a third feature matrix and a fourth feature matrix according to the first feature matrix and the second feature matrix; wherein the third feature matrix represents an updated first feature matrix; the fourth feature matrix represents an updated second feature matrix, and the third feature matrix and the fourth feature matrix are feature descriptions of the target object to be identified at different scales; performing feature fusion on the third feature matrix and the fourth feature matrix to determine a feature fusion result; identifying the target object to be identified in the frame image of the target object to be identified according to the feature fusion result; wherein the determining of the third feature matrix and the fourth feature matrix according to the first feature matrix and the second feature matrix comprises: determining a mask value according to the first feature matrix and the second feature matrix; performing first scale transformation on the first feature matrix and the second feature matrix according to the mask value to determine the third feature matrix, and the third feature matrix is determined by the first feature matrix and the second feature matrix together; performing second scale transformation on the first feature matrix and the second feature matrix according to the mask value to determine the fourth feature matrix, and the first scale transformation and the second scale transformation are multiplying the phases.

2. The method of claim 1, wherein, The determining of the third feature matrix according to the mask value, the first scale transformation on the first feature matrix and the second feature matrix comprises: performing spatial mapping conversion on the mask value to obtain a new mask value; performing first scale transformation on the first feature matrix and the second feature matrix according to the new mask value, a first parameter value and a second parameter value to obtain a first result value after first scale transformation; determining the third feature matrix according to the first result value.

3. The method of claim 1, wherein, The determining of the fourth feature matrix according to the mask value, the second scale transformation on the first feature matrix and the second feature matrix comprises: performing spatial mapping conversion on the mask value to obtain a new mask value; performing second scale transformation on the first feature matrix and the second feature matrix according to the new mask value, a first parameter value and a third parameter value to obtain a second result value after second scale transformation; determining the fourth feature matrix according to the second result value.

4. The method of claim 1, wherein, The feature fusion on the third feature matrix and the fourth feature matrix to determine the feature fusion result comprises: adjusting first weight information of the third feature matrix and second weight information of the fourth feature matrix; wherein the sum of the first weight information and the second weight information is 1; determining the feature fusion result according to the third feature matrix, the first weight information, the fourth feature matrix and the second weight information.

5. The method of claim 1, wherein, The target object to be identified is identified in the frame image of the target object to be identified according to the result of the feature fusion. The target object to be identified is labeled in the frame image of the target object to be identified according to the result of the feature fusion.

6. An object recognition apparatus characterized by comprising: The device comprises: A first determination unit is configured to acquire a frame image of a target object to be identified, and determine a first feature matrix and a second feature matrix in the frame image of the target object to be identified; wherein the first feature matrix represents semantic information; the second feature matrix represents resolution information; and the target object to be identified is an object in the frame image whose size ratio is less than a preset value. A second determination unit is configured to determine a third feature matrix and a fourth feature matrix according to the first feature matrix and the second feature matrix; wherein the third feature matrix represents an updated first feature matrix; and the fourth feature matrix represents an updated second feature matrix, and the third feature matrix and the fourth feature matrix are feature descriptions of the target object to be identified at different scales. A third determination unit is configured to perform feature fusion on the third feature matrix and the fourth feature matrix, and determine a result of the feature fusion. An identification unit is configured to identify the target object to be identified in the frame image of the target object to be identified according to the result of the feature fusion. The second determination unit is specifically configured to: Determine a mask value according to the first feature matrix and the second feature matrix. Perform first scale transformation on the first feature matrix and the second feature matrix according to the mask value, and determine a third feature matrix, wherein the third feature matrix is determined by the first feature matrix and the second feature matrix. Perform second scale transformation on the first feature matrix and the second feature matrix according to the mask value, and determine a fourth feature matrix, wherein the first scale transformation and the second scale transformation are multiplication of bit positions.

7. An electronic device, comprising: Comprise: A processor, and a memory connected to the processor in communication; The memory stores computer execution instructions; The processor executes the computer execution instructions stored in the memory to implement the method in any one of claims 1-5.

8. A computer-readable storage medium, characterized in that, The computer readable storage medium stores computer execution instructions, and the computer execution instructions are executed by the processor to implement the method in any one of claims 1-5.

9. A computer program product, characterised in that, A computer program is executed by the processor to implement the method in any one of claims 1-5.

Images