Handheld intelligent terminal, control method and device thereof and storage medium

Abstract

The invention belongs to the technical field of intelligent control, and discloses a handheld intelligent terminal control method and device, and a handheld intelligent terminal, and the method comprises the steps: obtaining the convolution parameters of each convolution layer in a preset convolution neural network for image classification and the equipment parameters of the handheld intelligent terminal, and carrying out the image classification according to the convolution parameters and the equipment parameters, the operation time of a preset convolutional neural network in the handheld intelligent terminal is predicted, if the operation time is smaller than the remaining operation time of the handheld intelligent terminal and the time difference between the operation time and the remaining operation time reaches a preset difference value, the handheld intelligent terminal is controlled to conduct convolution calculation on the signboard image through the preset convolutional neural network to obtain a feature map, and finally, according to a classification result of the feature map, determining whether the signboard of the power distribution network equipment is abnormal, and when it is determined that the remaining operation time of the handheld intelligent terminal is enough for operation of the preset convolutional neural network, detection of the signboard image can be executed, so that the detection success rate of the signboard of the power distribution network equipment is improved.

Description

Technical Field

[0001] The present invention belongs to the field of intelligent control technology, and specifically relates to a control method and device for a handheld intelligent terminal, a handheld intelligent terminal, and a storage medium. Background Art

[0002] With the rapid development of the power industry, distribution network equipment is widely deployed in various outdoor areas. Generally, each distribution network equipment is equipped with a signboard to serve as a reminder and early warning. Since the distribution network equipment is distributed in outdoor areas, the signboard is easily affected by outdoor environmental factors, resulting in blurred numbers, corrosion, damage, bending, etc., which leaves safety hazards for line operation, maintenance and management.

[0003] In order to ensure the safety of electricity use, it is necessary to conduct regular power inspections on distribution network equipment. One of the important tasks of the inspection is to check whether the identification plates of distribution network equipment are clear. The traditional manual inspection method is mainly based on high labor costs and low efficiency. In addition, it may not be able to accurately identify due to the personal ability of the inspectors.

[0004] Some image recognition methods based on machine vision are proposed in the prior art. The inspection personnel use handheld smart terminals to collect the image of the sign for recognition, which can improve the inspection efficiency and the objectivity of the inspection results.

[0005] However, in practice, it is found that in order to obtain higher accuracy, existing methods for signboard image recognition usually require a larger neural network structure and more parameters, which means that the amount of computation on the handheld smart terminal is relatively large. However, the processor and memory of the handheld smart terminal cannot always meet the needs of large computations, which makes the handheld smart terminal prone to jamming during execution, resulting in a low success rate for detecting distribution network equipment signboards. Summary of the invention

[0006] The object of the present invention is to provide a control method and device for a handheld intelligent terminal, a handheld intelligent terminal, and a storage medium, which can improve the detection success rate of identification plates of distribution network equipment.

[0007] The first aspect of the present invention discloses a control method for a handheld intelligent terminal, comprising:

[0008] Control the handheld intelligent terminal to collect the image of the sign of the distribution network equipment;

[0009] Obtaining convolution parameters of each convolution layer in a preset convolutional neural network and device parameters of the handheld intelligent terminal; wherein the preset convolutional neural network is used for image classification;

[0010] Predicting the operation time of the preset convolutional neural network in the handheld intelligent terminal according to the convolution parameters and the device parameters;

[0011] If the operation time is less than the remaining operation time of the handheld intelligent terminal, calculating the time difference between the operation time and the remaining operation time;

[0012] If the time difference reaches a preset difference, controlling the handheld intelligent terminal to perform convolution calculation on the sign image using the preset convolutional neural network to obtain a feature map;

[0013] According to the classification result of the characteristic graph, it is determined whether the identification plate of the distribution network equipment is abnormal.

[0014] In some embodiments, before controlling the handheld intelligent terminal to collect the image of the sign of the distribution network equipment, the method further includes:

[0015] Obtain target location information of the handheld smart terminal;

[0016] When the target location information matches the preset location information of the power distribution network equipment to be inspected, a prompt message is output, where the prompt message is used to remind the user to take a picture of the identification plate of the power distribution network equipment.

[0017] In some embodiments, obtaining the target location information of the handheld intelligent terminal includes:

[0018] Acquire multiple positioning sensing signals of a handheld smart terminal;

[0019] Extract features from the multiple positioning sensing signals, and identify and obtain multiple location information and the credibility of each location information; wherein the multiple location information corresponds to the multiple positioning sensing signals one by one;

[0020] Taking the position information with the highest credibility as the reference information, and calculating the distance between each position information other than the reference information and the reference information;

[0021] Filter out location information whose distance is greater than a preset distance from the multiple location information to obtain multiple candidate location information;

[0022] The information of multiple candidate locations is fused to obtain the target location information of the handheld smart terminal.

[0023] In some embodiments, the method further comprises:

[0024] If it is determined that the identification plate of the distribution network equipment is abnormal according to the classification result of the characteristic graph, a fault alarm message for replacing the identification plate is output;

[0025] If it is determined that the identification plate of the distribution network equipment is not abnormal according to the classification result of the characteristic graph, the clarity of the identification plate of the distribution network equipment is calculated. If the clarity is less than the specified clarity, a fault alarm message of replacing the identification plate is output.

[0026] A second aspect of the present invention discloses a control device for a handheld intelligent terminal, comprising:

[0027] A collection unit, used to control a handheld intelligent terminal to collect images of identification plates of distribution network equipment;

[0028] An acquisition unit, used to acquire convolution parameters of each convolution layer in a preset convolutional neural network and device parameters of the handheld intelligent terminal; wherein the preset convolutional neural network is used for image classification;

[0029] A prediction unit, configured to predict a computing time of the preset convolutional neural network in the handheld intelligent terminal according to the convolution parameters and the device parameters;

[0030] a calculation unit, configured to calculate a time difference between the calculation time and the remaining operation time if the calculation time is less than the remaining operation time of the handheld intelligent terminal;

[0031] A control unit, configured to control the handheld intelligent terminal to perform convolution calculation on the sign image using the preset convolutional neural network to obtain a feature map if the time difference reaches a preset difference;

[0032] A determination unit is used to determine whether the identification plate of the distribution network equipment is abnormal according to the classification result of the characteristic graph.

[0033] In some embodiments, the apparatus further comprises:

[0034] A positioning unit, used to obtain target position information of the handheld smart terminal before the acquisition unit controls the handheld smart terminal to acquire the image of the sign of the distribution network equipment;

[0035] The prompt unit is used to output prompt information when the target location information is consistent with the preset location information of the distribution network equipment to be inspected, wherein the prompt information is used to remind the user to take a picture of the identification plate of the distribution network equipment.

[0036] In some embodiments, the positioning unit includes:

[0037] An acquisition subunit, used to acquire multiple positioning perception signals of the handheld smart terminal before the acquisition unit controls the handheld smart terminal to acquire the image of the sign of the distribution network equipment;

[0038] An extraction subunit, used to extract features from the multiple positioning perception signals, and identify and obtain multiple position information and the credibility of each position information; wherein the multiple position information corresponds to the multiple positioning perception signals one by one;

[0039] A calculation subunit, configured to use the position information with the highest credibility as the reference information, and calculate the distance between each position information other than the reference information and the reference information;

[0040] A filtering subunit, used to filter out the location information whose distance is greater than a preset distance from the multiple location information, to obtain multiple candidate location information;

[0041] The fusion subunit is used to fuse multiple candidate location information to obtain the target location information of the handheld intelligent terminal.

[0042] In some embodiments, the apparatus further comprises:

[0043] An alarm unit is used to, after the determination unit determines whether the identification plate of the distribution network equipment is abnormal according to the classification result of the characteristic graph, output a fault alarm message for replacing the identification plate if it is determined that the identification plate of the distribution network equipment is abnormal according to the classification result of the characteristic graph; and, if it is determined that the identification plate of the distribution network equipment is not abnormal according to the classification result of the characteristic graph, calculate the clarity of the identification plate of the distribution network equipment, and output a fault alarm message for replacing the identification plate if the clarity is less than a specified clarity.

[0044] The third aspect of the present invention discloses a handheld intelligent terminal, comprising a memory storing executable program code and a processor coupled to the memory; the processor calls the executable program code stored in the memory to execute the control method of the handheld intelligent terminal disclosed in the first aspect.

[0045] A fourth aspect of the present invention discloses a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, wherein the computer program enables a computer to execute the control method of a handheld intelligent terminal disclosed in the first aspect.

[0046] The beneficial effect of the present invention lies in that by obtaining the convolution parameters of each convolution layer in the preset convolutional neural network for image classification and the device parameters of the handheld smart terminal, the operation time of the preset convolutional neural network in the handheld smart terminal is predicted according to the convolution parameters and the device parameters. If the operation time is less than the remaining operating time of the handheld smart terminal, and the time difference between the two reaches the preset difference, the handheld smart terminal is controlled to use the preset convolutional neural network to perform convolution calculation on the signboard image to obtain a feature map. Finally, according to the classification result of the feature map, it is determined whether the signboard of the distribution network equipment is abnormal. When it is determined that the remaining operating time of the handheld smart terminal is sufficient to supply the preset convolutional neural network operation, the signboard image detection can be performed, thereby improving the detection success rate of the distribution network equipment signboard. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] The accompanying drawings herein show specific examples of the technical solutions described in the present invention, and together with the specific implementation methods, constitute a part of the specification, and are used to explain the technical solutions, principles and effects of the present invention.

[0048] Unless otherwise specified or defined, the same figure marks in different drawings represent the same or similar technical features, and the same or similar technical features may also be represented by different figure marks.

[0049] Figure 1 is a flow chart of a control method of a handheld intelligent terminal disclosed in an embodiment of the present invention;

[0050] Figure 2 It is a structural schematic diagram of a control device of a handheld intelligent terminal disclosed in an embodiment of the present invention;

[0051] Figure 3 It is a structural schematic diagram of a handheld intelligent terminal disclosed in an embodiment of the present invention.

[0052] Description of reference numerals:

[0053] 201, acquisition unit; 202, acquisition unit; 203, prediction unit; 204, calculation unit; 205, control unit; 206, determination unit; 301, memory; 302, processor. DETAILED DESCRIPTION

[0054] In order to facilitate the understanding of the present invention, specific embodiments of the present invention will be described in more detail below with reference to the accompanying drawings.

[0055] Unless otherwise specified or defined, all technical and scientific terms used herein have the same meaning as those generally understood by those skilled in the art. In the case of combining the technical solution of the present invention with realistic scenarios, all technical and scientific terms used herein may also have meanings corresponding to the purpose of implementing the technical solution of the present invention. The "first, second..." used herein is only used to distinguish the names and does not represent a specific quantity or order. The term "and / or" used herein includes any and all combinations of one or more related listed items.

[0056] It should be noted that when a component is considered to be "fixed to" another component, it can be directly fixed to the other component or there can be a central component; when an component is considered to be "connected to" another component, it can be directly connected to the other component or there can be a central component at the same time; when an component is considered to be "installed on" another component, it can be directly installed on the other component or there can be a central component at the same time. When an component is considered to be "set on" another component, it can be directly set on the other component or there can be a central component at the same time.

[0057] Unless otherwise specified or defined, the "said" and "the" used in this document refer to the technical features or technical contents mentioned or described before the corresponding position, and the technical features or technical contents may be the same as or similar to the technical features or technical contents mentioned therein. In addition, the terms "including" and "having" and any variations thereof used in this document are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units that are not listed, or may optionally include other steps or units inherent to these processes, methods, products or devices.

[0058] The embodiment of the present invention discloses a control method for a handheld intelligent terminal. The execution subject of the method may be a handheld intelligent terminal such as a computer, a laptop computer, a tablet computer, or a control device embedded in the handheld intelligent terminal, and the present invention does not limit this.

[0059] like Figure 1 As shown, in this embodiment, the method includes the following steps S10 to S60:

[0060] S10. Control the handheld intelligent terminal to collect the image of the identification plate of the distribution network equipment.

[0061] In an embodiment of the present invention, before executing step S10, the target location information of the handheld intelligent terminal can be obtained first, and when the target location information matches the preset location information of the power distribution network equipment to be inspected, a prompt message is output, and the prompt message is used to remind the user to take a picture of the identification plate of the power distribution network equipment. After the prompt message is output, when a shooting instruction input by the user is received, step S10 is executed.

[0062] As a preferred implementation, the method for obtaining the target location information of a handheld smart terminal can be: obtaining multiple positioning sensing signals of the handheld smart terminal, performing feature extraction on the multiple positioning sensing signals, and identifying and obtaining multiple location information and the credibility of each location information; wherein the multiple location information correspond one-to-one to the multiple positioning sensing signals; using the location information with the greatest credibility as the benchmark information, and calculating the distance between each location information other than the benchmark information and the benchmark information; filtering out the location information whose distance is greater than a preset distance from the multiple location information to obtain multiple candidate location information; finally, fusing the multiple candidate location information to obtain the target location information of the handheld smart terminal.

[0063] This preferred implementation, by fusing multiple high-reliability positioning perception signals, can improve positioning accuracy compared to positioning using a single signal source.

[0064] S20, obtaining convolution parameters of each convolution layer in a preset convolution neural network and device parameters of the handheld intelligent terminal. The preset convolution neural network is used for image classification.

[0065] In an embodiment of the present invention, when it is necessary to deploy a preset convolutional neural network in a handheld smart terminal for operation, the handheld smart terminal can identify each convolutional layer in the preset convolutional neural network and extract the convolution parameters of each convolutional layer, wherein the convolution parameters may include the number of input channels Ci, height H and width W and the number of output channels Co, height KH and width KW of the convolution kernel, etc.

[0066] At the same time, the handheld smart terminal can also obtain the device parameters of the handheld smart terminal, for example, the device parameters of the handheld smart terminal can be obtained by directly reading the attribute information attached to the handheld smart terminal, the attribute information can include the device parameters of the handheld smart terminal, or the device parameters of the handheld smart terminal can be obtained by running a test on the handheld smart terminal, which is not limited in the embodiment of the present invention. Among them, the device parameters may include the peak memory bandwidth and the maximum number of operations per second of the handheld smart terminal.

[0067] In an embodiment of the present invention, the preset convolutional neural network may include two parts, namely a target detection model and a classification model. The training process of the target detection model may include:

[0068] Collect sample images of the signboards of distribution network equipment under various weather and light conditions and mark the target frames; after data preprocessing of the sample images, use the YOLOv3 network as the target detection network, and perform model building, model training and model tuning in sequence to complete the training and obtain the target detection model; input the sample images to be detected into the trained target detection model to obtain the prediction results of the signboard location area in the current sample image. The present invention uses the YOLOv3 network as the target detection network, which can realize the accurate recognition of the signboards of distribution network equipment in all-weather scenarios and has good generalization ability.

[0069] In addition, the classification model can use the Faster R-CNN deep learning network, and the network structure includes a convolutional network, an extraction network, and a Fast R-CNN detection network. The convolutional network can use the neural structure search network NasNet, which contains 49 convolutional layers. Except for the first layer, which is a 7*7 convolution, the rest are 1*1 convolution and 3*3 convolution. After the convolutional network performs deep feature extraction on the input image, a feature map is generated. Then, the extraction network continues to use 3*3 convolution on the feature map generated by the convolutional network to generate a target feature map with the same size and depth, and calculates the confidence and the position coordinates of the sign position area through two parallel 1*1 convolutional layers to generate a sign area map. Finally, the Fast R-CNN detection network performs a region of interest pooling operation on the sign area map based on the sign area map generated by the extraction network, and then calculates the confidence of the target type and the target position coordinates through two parallel sets of fully connected layers.

[0070] S30: predicting the operation time of a preset convolutional neural network in the handheld intelligent terminal according to the convolution parameters and the device parameters.

[0071] For each convolution layer in the preset convolutional neural network, the operator time of each preset acceleration algorithm for performing convolution operation on the convolution layer in the handheld intelligent terminal can be estimated according to the convolution parameters and device parameters of the convolution layer. For example, the operator time t1 of the first convolution layer performed by the preset acceleration algorithm A in the handheld intelligent terminal can be estimated according to the convolution parameter a of the first convolution layer and the device parameters; the operator time t1 of the first convolution layer performed by the preset acceleration algorithm B in the handheld intelligent terminal can be estimated according to the convolution parameter a of the first convolution layer and the device parameters; the operator time t2 of the second convolution layer performed by the preset acceleration algorithm A in the handheld intelligent terminal can be estimated according to the convolution parameter b of the second convolution layer and the device parameters; the operator time t2 of the second convolution layer performed by the preset acceleration algorithm B in the handheld intelligent terminal can be estimated according to the convolution parameter b of the second convolution layer and the device parameters, and the like. Finally, the operation time of each convolutional layer in the convolutional neural network is summed up to obtain the operation time of the preset convolutional neural network in the handheld intelligent terminal.

[0072] S40: If the operation time is less than the remaining operation time of the handheld intelligent terminal, calculate the time difference between the operation time and the remaining operation time.

[0073] Among them, the remaining operating time of the handheld smart terminal refers to the time from the current moment to the automatic shutdown due to exhaustion of power or automatic shutdown due to freezing. If the operation time is less than the remaining operating time of the handheld smart terminal, it means that the remaining operating time of the handheld smart terminal is sufficient to supply the preset convolutional neural network operation. And, further, by calculating the time difference between the operation time and the remaining operating time, the redundancy of the handheld smart terminal after the operation can be predicted. If the time difference is less than the preset difference, it means that the redundancy of the handheld smart terminal after the operation is insufficient, and the convolution calculation of the sign image is not performed.

[0074] S50: If the time difference reaches a preset difference, control the handheld intelligent terminal to use a preset convolutional neural network to perform convolution calculation on the sign image to obtain a feature map.

[0075] If the remaining running time of the handheld smart terminal is sufficient to supply the preset convolutional neural network operation, and the time difference reaches the preset difference, it means that the redundancy of the handheld smart terminal after the operation is sufficient, then the convolution calculation of the sign image is performed.

[0076] S60. Determine whether the identification plate of the distribution network equipment is abnormal based on the classification result of the characteristic graph.

[0077] If it is determined based on the classification results of the characteristic graph that the nameplate of the distribution network equipment is abnormal, a fault alarm message for replacing the nameplate can be output; if it is determined based on the classification results of the characteristic graph that the nameplate of the distribution network equipment is not abnormal, the clarity of the nameplate of the distribution network equipment can be further calculated. If the clarity is less than the specified clarity, it means that the nameplate image is not clear, and a fault alarm message for replacing the nameplate can also be output.

[0078] like Figure 2 As shown, the embodiment of the present invention discloses a control device for a handheld intelligent terminal, including a collection unit 201, an acquisition unit 202, a prediction unit 203, a calculation unit 204, a control unit 205, and a determination unit 206, wherein:

[0079] The collection unit 201 is used to control the handheld intelligent terminal to collect the image of the sign of the distribution network equipment;

[0080] The acquisition unit 202 is used to acquire the convolution parameters of each convolution layer in the preset convolution neural network and the device parameters of the handheld intelligent terminal; wherein the preset convolution neural network is used for image classification;

[0081] The prediction unit 203 is used to predict the operation time of the preset convolutional neural network in the handheld intelligent terminal according to the convolution parameters and the device parameters;

[0082] The calculation unit 204 is used to calculate the time difference between the calculation time and the remaining operation time if the calculation time is less than the remaining operation time of the handheld intelligent terminal;

[0083] The control unit 205 is used to control the handheld intelligent terminal to perform convolution calculation on the sign image using a preset convolutional neural network to obtain a feature map if the time difference reaches a preset difference;

[0084] The determination unit 206 is used to determine whether the identification plate of the distribution network equipment is abnormal according to the classification result of the characteristic graph.

[0085] As an optional implementation, the control device of the handheld intelligent terminal may further include the following units not shown in the figure:

[0086] A positioning unit, used to obtain target position information of the handheld smart terminal before the acquisition unit 201 controls the handheld smart terminal to acquire the image of the sign of the distribution network equipment;

[0087] The prompt unit is used to output prompt information when the target location information matches the preset location information of the distribution network equipment to be inspected, wherein the prompt information is used to remind the user to take a picture of the identification plate of the distribution network equipment.

[0088] As an optional implementation, the positioning unit may include the following subunits (not shown):

[0089] An acquisition subunit, used to acquire multiple positioning sensing signals of the handheld smart terminal before the acquisition unit 201 controls the handheld smart terminal to acquire the image of the sign of the distribution network equipment;

[0090] An extraction subunit, used to extract features from the multiple positioning perception signals, and identify and obtain multiple position information and the credibility of each position information; wherein the multiple position information corresponds to the multiple positioning perception signals one by one;

[0091] A calculation subunit, used to use the position information with the highest credibility as the reference information, and calculate the distance between each position information other than the reference information and the reference information;

[0092] A filtering subunit, used to filter out the location information whose distance is greater than a preset distance from the multiple location information, to obtain multiple candidate location information;

[0093] The fusion subunit is used to fuse multiple candidate location information to obtain the target location information of the handheld intelligent terminal.

[0094] As an optional implementation, the control device of the handheld intelligent terminal may further include the following units not shown in the figure:

[0095] An alarm unit is used for, after the determination unit 206 determines whether the nameplate of the distribution network equipment is abnormal according to the classification result of the characteristic graph, if it is determined according to the classification result of the characteristic graph that the nameplate of the distribution network equipment is abnormal, outputting a fault alarm message for replacing the nameplate; and, if it is determined according to the classification result of the characteristic graph that the nameplate of the distribution network equipment is not abnormal, calculating the clarity of the nameplate of the distribution network equipment, and outputting a fault alarm message for replacing the nameplate if the clarity is less than a specified clarity.

[0096] like Figure 3 As shown, the embodiment of the present invention discloses a handheld intelligent terminal, including a memory 301 storing executable program codes and a processor 302 coupled to the memory 301;

[0097] The processor 302 calls the executable program code stored in the memory 301 to execute the control method of the handheld intelligent terminal described in the above embodiments.

[0098] The embodiment of the present invention further discloses a computer-readable storage medium storing a computer program, wherein the computer program enables a computer to execute the control method of the handheld intelligent terminal described in the above embodiments.

[0099] The purpose of the above embodiments is to exemplarily reproduce and deduce the technical solution of the present invention, and to fully describe the technical solution, purpose and effect of the present invention. Its purpose is to make the public understand the disclosed content of the present invention more thoroughly and comprehensively, and it does not limit the protection scope of the present invention.

[0100] The above embodiments are not exhaustive enumerations of the present invention, and there may be multiple other implementations not listed. Any replacement and improvement made without violating the concept of the present invention shall fall within the protection scope of the present invention.

Claims

1. A control method for a handheld intelligent terminal, characterized in that: include: Control the handheld intelligent terminal to collect the image of the sign of the distribution network equipment; Obtaining convolution parameters of each convolution layer in a preset convolutional neural network and device parameters of the handheld intelligent terminal; wherein the preset convolutional neural network is used for image classification; Predicting the operation time of the preset convolutional neural network in the handheld intelligent terminal according to the convolution parameters and the device parameters; If the operation time is less than the remaining operation time of the handheld intelligent terminal, calculating the time difference between the operation time and the remaining operation time; If the time difference reaches a preset difference, controlling the handheld intelligent terminal to perform convolution calculation on the sign image using the preset convolutional neural network to obtain a feature map; According to the classification result of the characteristic graph, it is determined whether the identification plate of the distribution network equipment is abnormal.

2. The control method of the handheld intelligent terminal according to claim 1, characterized in that: Before controlling the handheld intelligent terminal to collect the image of the sign of the distribution network equipment, the method further includes: Obtain target location information of the handheld smart terminal; When the target location information matches the preset location information of the power distribution network equipment to be inspected, a prompt message is output, where the prompt message is used to remind the user to take a picture of the identification plate of the power distribution network equipment.

3. The control method of the handheld intelligent terminal according to claim 2, characterized in that: The step of obtaining the target location information of the handheld intelligent terminal includes: Acquire multiple positioning sensing signals of a handheld smart terminal; Extract features from the multiple positioning sensing signals, and identify and obtain multiple location information and the credibility of each location information; wherein the multiple location information corresponds to the multiple positioning sensing signals one by one; Taking the position information with the highest credibility as the reference information, and calculating the distance between each position information other than the reference information and the reference information; Filter out location information whose distance is greater than a preset distance from the multiple location information to obtain multiple candidate location information; The information of multiple candidate locations is fused to obtain the target location information of the handheld smart terminal.

4. The control method of a handheld intelligent terminal according to any one of claims 1 to 3, characterized in that: The method further comprises: If it is determined that the identification plate of the distribution network equipment is abnormal according to the classification result of the characteristic graph, a fault alarm message for replacing the identification plate is output; If it is determined that the identification plate of the distribution network equipment is not abnormal according to the classification result of the characteristic graph, the clarity of the identification plate of the distribution network equipment is calculated. If the clarity is less than the specified clarity, a fault alarm message of replacing the identification plate is output.

5. A control device for a handheld intelligent terminal, characterized in that: include: A collection unit, used to control a handheld intelligent terminal to collect images of identification plates of distribution network equipment; An acquisition unit, used to acquire convolution parameters of each convolution layer in a preset convolutional neural network and device parameters of the handheld intelligent terminal; wherein the preset convolutional neural network is used for image classification; A prediction unit, configured to predict a computing time of the preset convolutional neural network in the handheld intelligent terminal according to the convolution parameters and the device parameters; a calculation unit, configured to calculate a time difference between the calculation time and the remaining operation time if the calculation time is less than the remaining operation time of the handheld intelligent terminal; A control unit, configured to control the handheld intelligent terminal to perform convolution calculation on the sign image using the preset convolutional neural network to obtain a feature map if the time difference reaches a preset difference; A determination unit is used to determine whether the identification plate of the distribution network equipment is abnormal according to the classification result of the characteristic graph.

6. The control device of the handheld intelligent terminal according to claim 5, characterized in that: The device also includes: A positioning unit, used to obtain target position information of the handheld smart terminal before the acquisition unit controls the handheld smart terminal to acquire the image of the sign of the distribution network equipment; The prompt unit is used to output prompt information when the target location information is consistent with the preset location information of the distribution network equipment to be inspected, wherein the prompt information is used to remind the user to take a picture of the identification plate of the distribution network equipment.

7. The control device of the handheld intelligent terminal according to claim 6, characterized in that: The positioning unit comprises: An acquisition subunit, used to acquire multiple positioning perception signals of the handheld smart terminal before the acquisition unit controls the handheld smart terminal to acquire the image of the sign of the distribution network equipment; An extraction subunit, used to extract features from the multiple positioning perception signals, and identify and obtain multiple position information and the credibility of each position information; wherein the multiple position information corresponds to the multiple positioning perception signals one by one; A calculation subunit, configured to use the position information with the highest credibility as the reference information, and calculate the distance between each position information other than the reference information and the reference information; A filtering subunit, used to filter out the location information whose distance is greater than a preset distance from the multiple location information, to obtain multiple candidate location information; The fusion subunit is used to fuse multiple candidate location information to obtain the target location information of the handheld intelligent terminal.

8. The control device for a handheld intelligent terminal according to any one of claims 5 to 7, characterized in that: The device also includes: An alarm unit is used to, after the determination unit determines whether the identification plate of the distribution network equipment is abnormal according to the classification result of the characteristic graph, output a fault alarm message for replacing the identification plate if it is determined that the identification plate of the distribution network equipment is abnormal according to the classification result of the characteristic graph; and, if it is determined that the identification plate of the distribution network equipment is not abnormal according to the classification result of the characteristic graph, calculate the clarity of the identification plate of the distribution network equipment, and output a fault alarm message for replacing the identification plate if the clarity is less than a specified clarity.

9. A handheld intelligent terminal, characterized in that: It comprises a memory storing executable program code and a processor coupled to the memory; the processor calls the executable program code stored in the memory to execute the control method of the handheld intelligent terminal according to any one of claims 1 to 4.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, wherein the computer program enables a computer to execute the control method of a handheld intelligent terminal according to any one of claims 1 to 4.

Images