Construction waste resourceful treatment intelligent sorting method and system

Abstract

The invention provides an intelligent sorting method and system for resourceful treatment of construction waste, and particularly relates to the technical field of construction waste recycling The method comprises the steps that collision audio of a target raw material is obtained, the target raw material is to-be-sorted construction waste, and the collision audio of the target raw material is obtained; the collision audio is audio generated when the target raw material collides with a preset obstacle in the transmission process; the collision audio is analyzed, probability data and category data are obtained, the probability data is used for representing the probability that the target raw materials include the target materials, and the category data is used for representing the material categories of the target materials included in the target raw materials; the sorting robot is controlled to conduct visual sorting treatment on the target raw materials according to the analysis result; the higher the probability represented by the probability data is, the higher the image acquisition frequency of the target raw material is, and the feature library used by the sorting robot in the detection process is determined based on the category data. According to the invention, the accuracy of the recognition result of the construction waste can be improved, and the economic benefit of sorting and recycling the construction waste can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of construction waste recycling, and particularly relates to an intelligent sorting method and system for resource treatment of construction waste. Background Art

[0002] With the growth of the global population and the acceleration of the urbanization process, more construction waste is generated. Some materials have potential value in construction and are easily reusable and recyclable, such as various metal materials or plastic materials. By effectively sorting construction waste, the treatment cost of construction waste can be effectively reduced and environmental pollution can be reduced.

[0003] With the popularization and application of computer vision technology, the sorting efficiency of construction waste has been effectively improved. However, at present, the accuracy rate of the recognition results of computer vision technology is relatively low, and materials with high recycling value such as metals cannot be fully sorted, which makes the economic benefits of construction waste sorting and recycling not as expected. Summary of the Invention

[0004] The purpose of the present disclosure is to provide an intelligent sorting method and system for resource treatment of construction waste, which is used to solve the technical problems existing in the existing intelligent sorting scheme of construction waste, including the low accuracy rate of the recognition results of construction waste and the low economic benefits of sorting and recycling.

[0005] In a first aspect, an embodiment of the present invention provides an intelligent sorting method for resource treatment of construction waste, and the method includes: Obtain the collision audio of the target raw material, where the target raw material is the construction waste to be sorted, and the collision audio is the audio generated by the target raw material colliding with a preset obstacle during the transmission process; Analyze the collision audio to obtain an analysis result, where the analysis result includes probability data and category data. The probability data is used to represent the probability that the target raw material includes the target material, and the category data is used to represent the material category of the target material included in the target raw material. The target material is one of a preset variety of high-value materials; Control a sorting robot to perform visual sorting processing on the target raw material according to the analysis result, where the sorting robot is used to: collect an image of the target raw material, and grab the target material to a target position when it is detected that the image includes the target material; The higher the probability represented by the probability data, the higher the image acquisition frequency of the target raw material, and the feature library used by the sorting robot during the detection process is determined based on the category data.

[0006] In one embodiment, analyzing the collision audio to obtain an analysis result includes: Extracting features from the collision audio to obtain a first audio feature; Performing similarity calculations between each of a plurality of second audio features and the first audio feature to obtain a plurality of similarity values, where the plurality of second audio features correspond one-to-one to a plurality of high-value materials, and the second audio feature is used to represent the characteristics of the audio generated when the corresponding high-value material collides with the preset obstacle; the similarity value is used to represent the feature similarity between the corresponding second audio feature and the first audio feature; Determining the maximum similarity value among the plurality of similarity values as the target similarity value, and generating the analysis result based on the target similarity value and the high-value material corresponding to the target similarity value, where the probability data is determined based on the target similarity value, and the category data is determined based on the high-value material corresponding to the target similarity value.

[0007] In one embodiment, the collision audio is obtained based on an audio acquisition device, and the audio sensing end of the audio acquisition device is pressed tightly against the outer wall of the preset obstacle.

[0008] In one embodiment, the target raw material is transported by a conveyor belt, and the forward direction of the target raw material in the conveyor belt is the first direction; The conveyor belt includes a transmission section and a sorting section connected along the first direction, where the maximum length of the transmission section in the second direction is less than the minimum length of the sorting section in the second direction, and the length of the sorting section in the second direction gradually increases along the first direction, the second direction is perpendicular to the gravity direction, and the second direction is perpendicular to the first direction; The sorting robot performs visual sorting processing on the target raw material in the sorting section.

[0009] In one embodiment, the sorting section is inclined, and the maximum height of the sorting section in the gravity direction is less than the height of the transmission section in the gravity direction, and the preset obstacle is arranged between the sorting section and the transmission section.

[0010] In a second aspect, an intelligent sorting system for resource treatment of construction waste according to an embodiment of the present invention further includes: An audio acquisition module, configured to acquire the collision audio of the target raw material, where the target raw material is construction waste to be sorted, and the collision audio is the audio generated when the target raw material collides with a preset obstacle during transportation; An audio analysis module is configured to analyze the collision audio to obtain an analysis result. The analysis result includes probability data and category data. The probability data is used to represent the probability that the target raw material includes the target material, and the category data is used to represent the material category of the target material included in the target raw material. The target material is one of a plurality of preset high-value materials. A sorting processing module is configured to control a sorting robot to perform visual sorting processing on the target raw material according to the analysis result. The sorting robot is configured to: collect an image of the target raw material, and when it is detected that the image includes the target material, grab the target material to a target position. The higher the probability represented by the probability data, the higher the image acquisition frequency of the target raw material. The feature library used by the sorting robot in the detection process is determined based on the category data.

[0011] In one embodiment, the audio analysis module is specifically configured to: Extract features from the collision audio to obtain a first audio feature; Perform similarity calculation between each of a plurality of second audio features and the first audio feature to obtain a plurality of similarity values. The plurality of second audio features correspond one-to-one to the plurality of high-value materials. The second audio feature is used to represent the characteristics of the audio generated when the corresponding high-value material collides with the preset obstacle. The similarity value is used to represent the feature similarity between the corresponding second audio feature and the first audio feature. Determine the maximum similarity value among the plurality of similarity values as the target similarity value, and generate the analysis result according to the target similarity value and the high-value material corresponding to the target similarity value. The probability data is determined based on the target similarity value, and the category data is determined based on the high-value material corresponding to the target similarity value.

[0012] In one embodiment, the collision audio is obtained by an audio acquisition device, and the audio sensing end of the audio acquisition device is tightly pressed against the outer wall of the preset obstacle.

[0013] In one embodiment, the target raw material is transported by a conveyor belt, and the forward direction of the target raw material on the conveyor belt is the first direction; The conveyor belt includes a transmission section and a sorting section connected along the first direction. The maximum length of the transmission section in the second direction is less than the minimum length of the sorting section in the second direction, and the length of the sorting section in the second direction gradually increases along the first direction. The second direction is perpendicular to the gravity direction and perpendicular to the first direction. The sorting robot performs visual sorting on the target raw material in the sorting section.

[0014] In one embodiment, the sorting section is inclined, and the maximum height of the sorting section in the direction of gravity is less than the height of the transmission section in the direction of gravity. The preset obstacle is arranged between the sorting section and the transmission section.

[0015] In a third aspect, an embodiment of the present invention further provides an electronic device, including a processor, a memory, and a computer program stored on the memory and executable on the processor. When the computer program is executed by the processor, the steps of the above intelligent sorting method for construction waste resource treatment are implemented.

[0016] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above intelligent sorting method for construction waste resource treatment are implemented.

[0017] In a fifth aspect, the present disclosure provides a computer program product, including computer instructions. When the computer instructions are executed by a processor, the steps of the above intelligent sorting method for construction waste resource treatment are implemented.

[0018] In the embodiment of the present invention, through the setting of the preset obstacle, it actively collides with the construction waste to be sorted, and collects the collision audio generated when the two collide. Then, the collision audio is analyzed to determine the probability that high-value materials exist in the construction waste to be sorted, and the categories of the existing high-value materials. Then, the image acquisition frequency and the feature library used by the sorting robot during subsequent visual sorting are determined, which can enable the sorting robot to dynamically configure appropriate image acquisition frequencies and visual feature libraries for different construction wastes, thereby improving the recognition accuracy of the sorting robot for high-value materials in construction waste, enhancing the sorting effect of the sorting robot for high-value materials existing in construction waste, helping the sorting robot to grab more high-value materials from construction waste, and thus improving the economic benefits of sorting and recycling construction waste. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 is a schematic flowchart of an intelligent sorting method for construction waste resource treatment provided by an embodiment of the present invention; Figure 2 is a schematic structural diagram of an intelligent sorting system for construction waste resource treatment provided by an embodiment of the present invention; Figure 3 is a schematic diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0020] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0021] An intelligent sorting method for resource treatment of construction waste is provided in an embodiment of the present invention. Refer to Figure 1 , Figure 1 which is a flowchart of the intelligent sorting method for resource treatment of construction waste provided in the embodiment of the present invention. As Figure 1 shown, it includes the following steps: Step 101, obtain the collision audio of the target raw material.

[0022] Among them, the target raw material is the construction waste to be sorted, and the collision audio is the audio generated when the target raw material collides with a preset obstacle during the transmission process.

[0023] The preset obstacle is arranged on the transmission path of the target raw material to ensure that the preset obstacle contacts and collides with the target raw material.

[0024] Exemplarily, the preset obstacle can be columnar, conical, etc., so as to reduce the probability of the preset obstacle hindering the transmission of the target raw material while colliding with the target raw material to generate collision audio.

[0025] Step 102, analyze the audio data to obtain an analysis result.

[0026] Among them, the analysis result includes probability data and category data. The probability data is used to represent the probability that the target raw material includes the target material, and the category data is used to represent the material category of the target material included in the target raw material. The target material is one of a variety of preset high-value materials.

[0027] The aforementioned high-value materials can be understood as materials with high recycling value in construction waste, such as: steel, aluminum, copper cable with skin, copper wire without skin, special glass, PVC profiles, rosewood waste, LED drive power supplies, etc.

[0028] Among them, the materials with high recycling value can be understood as materials with a recycling unit price greater than or equal to the set unit price threshold. For example, the unit price threshold can be set to 500 yuan per ton. In application, users can adaptively adjust the unit price threshold based on actual needs, such as: 600 yuan per ton, 850 yuan per ton, etc. The present invention does not limit the specific value of the unit price threshold.

[0029] Step 103: Control the sorting robot to perform visual sorting on the target raw material according to the analysis result.

[0030] Among them, the sorting robot is used to: collect an image of the target raw material, and when it detects that the image includes the target material, grab the target material to the target position.

[0031] The higher the probability represented by the probability data, the higher the image acquisition frequency of the target raw material. The feature library used by the sorting robot during the detection process is determined based on the category data.

[0032] Specifically, the process of controlling the sorting robot to perform visual sorting on the target raw material according to the analysis result can be as follows: Set the image acquisition frequency of the sorting robot for the target raw material according to the probability data included in the analysis result, and set the feature library used by the sorting robot during the detection process according to the category data included in the analysis result; Based on the foregoing image acquisition frequency, control the sorting robot to take multiple shots of the target raw material to obtain multiple sampled images of the target raw material (the image capture times of the multiple sampled images are different); Then, perform image feature extraction on the multiple sampled images respectively to obtain multiple sampled image features corresponding to the multiple sampled images one by one; For the target sampled image feature among the multiple sampled image features, perform similarity calculation between the target sampled image feature and multiple reference features in the foregoing feature library to obtain multiple image similarity values corresponding to the target sampled image feature. Among them, the multiple image similarity values corresponding to the target sampled image feature correspond to the multiple reference features one by one. The image similarity value is used to represent the image feature similarity between the target sampled image feature and the corresponding reference feature. The target sampled image feature can be understood as any one of the multiple sampled image features; When the multiple image similarity values corresponding to the target sampled image feature include an image similarity value greater than a preset image threshold, it is determined that the sampled image corresponding to the target sampled image feature includes the target material. Based on the sampling position of the sampled image corresponding to the target sampled image feature, grab the target material, and move the target material to the target position after grabbing the target material.

[0033] For example, the image acquisition frequency can be one image per second, one image per three seconds, etc. The foregoing multiple high-value materials correspond to multiple different material categories, and multiple different material categories respectively correspond to multiple different feature libraries. Each feature library is used to represent the image features of high-value materials corresponding to the corresponding material category.

[0034] Exemplarily, algorithms such as SIFT, HOG, and ORB can be applied to complete the above-mentioned operation of image feature extraction. It should be understood that the reference features included in each feature library are also extracted based on the same algorithm.

[0035] Moreover, algorithms such as Manhattan distance, Euclidean distance, and cosine similarity can be applied to complete the calculation of the above-mentioned image similarity value.

[0036] It should be noted that grasping the target material based on the sampling position of the sampling image corresponding to the target sampling image feature should be understood as: When the target raw material is temporarily stagnant when transported to the area where the sorting robot is located, grasping the target material based on the sampling position of the sampling image corresponding to the target sampling image feature is sufficient; When the target raw material is not temporarily stagnant when transported to the area where the sorting robot is located, then based on the sampling position of the sampling image corresponding to the target sampling image feature, the sampling time of the sampling image corresponding to the target sampling image feature, and the transmission rate of the target raw material, predict the position of the sampling image corresponding to the target sampling image feature at the current moment, and grasp the target material at the position of the sampling image corresponding to the target sampling image feature at the current moment.

[0037] Among them, the target position can be the position for temporarily storing the target material. For example, it can be the position where the movable storage bin is located; it can also be the position of the production line for performing refined recycling processing on the target material.

[0038] In the embodiment of the present invention, through the setting of preset obstacles, actively collide with the construction waste to be sorted, and collect the collision audio generated when the two collide, and then analyze the collision audio to determine the probability that high-value materials exist in the construction waste to be sorted, and the categories of the existing high-value materials, and then determine the image acquisition frequency and the feature library used by the sorting robot in subsequent visual sorting processing, which can enable the sorting robot to dynamically configure appropriate image acquisition frequencies and visual feature libraries for different construction wastes, thereby improving the recognition accuracy of high-value materials in construction waste by the sorting robot, enhancing the sorting effect of high-value materials existing in construction waste by the sorting robot, so as to help the sorting robot grab more high-value materials from construction waste, thereby improving the economic benefits of sorting and recycling construction waste.

[0039] In one embodiment, the analysis of the collision audio to obtain an analysis result includes: Extract features from the collision audio to obtain the first audio feature; Calculate the similarity between each of the multiple second audio features and the first audio feature to obtain multiple similarity values. Among them, the multiple second audio features correspond to the multiple high-value materials one by one. The second audio feature is used to represent the characteristics of the audio generated when the corresponding high-value material collides with the preset obstacle. The similarity value is used to represent the feature similarity between the corresponding second audio feature and the first audio feature. Determine the maximum similarity value among the multiple similarity values as the target similarity value, and generate the analysis result based on the target similarity value and the high-value material corresponding to the target similarity value. Among them, the probability data is determined based on the target similarity value, and the category data is determined based on the high-value material corresponding to the target similarity value.

[0040] Exemplarily, the feature extraction of the collision audio can be completed by time-frequency domain conversion. For example, methods such as short-time Fourier transform or wavelet transform can be used, or the method of linear predictive coding can be used. Also, the feature extraction of the collision audio can be completed by a trained convolutional neural network.

[0041] The calculation of the similarity value can be completed by the method of dynamic time warping, or based on deep learning, or by the method of cosine similarity.

[0042] In this embodiment, by comparing audio features, the type of high-value material corresponding to the collision audio can be identified conveniently and accurately, so as to ensure the operation accuracy of subsequent visual sorting processing, and further ensure the sorting effect of the high-value materials existing in the construction waste by the sorting robot.

[0043] In one embodiment, the collision audio is obtained based on an audio acquisition device, and the audio sensing end of the audio acquisition device is tightly pressed against the outer wall of the preset obstacle.

[0044] In the case of a large amount of noise in the transmission process of the target raw material, by tightly pressing the audio sensing end of the audio acquisition device for collecting the collision audio against the outer wall of the preset obstacle, the preset obstacle is used as a solid audio propagation medium to replace the conventional gaseous audio propagation medium (i.e., air). On the one hand, this can reduce the acquisition delay of the collision audio and improve the subsequent detection and sorting efficiency of the target raw material. On the other hand, it can reduce the mixing of noise and improve the accuracy of the analysis result of the subsequent analysis of the collision audio.

[0045] Exemplarily, the audio acquisition device can be an acceleration sensor, an acoustic emission sensor, etc.

[0046] In one embodiment, the target raw material is transported by a conveyor belt, and the forward direction of the target raw material in the conveyor belt is the first direction; The conveyor belt includes a transport section and a sorting section connected along the first direction. Among them, the maximum length of the transport section in the second direction is less than the minimum length of the sorting section in the second direction, and the length of the sorting section in the second direction gradually increases along the first direction. The second direction is perpendicular to the direction of gravity, and the second direction is perpendicular to the first direction; The sorting robot performs visual sorting processing on the target raw material in the sorting section.

[0047] In this embodiment, through the setting of the sorting section, when the target raw material is transported to the area where the sorting robot is located, it shows a diffusing trend, so as to inhibit the stacking of the target raw material, and further reduce the occurrence probability of the problem of missed recognition of the target material caused by stacking, enabling the sorting robot to perform more accurate and efficient visual sorting processing.

[0048] Among them, if the first direction is set as the length direction, the second direction can be understood as the width direction. The length of the transport section in the second direction can be understood as the distance between the two side walls of the transport section, and the transport path of the target raw material is located between the two side walls of the transport section; similarly, the length of the sorting section in the second direction can be understood as the distance between the two side walls of the sorting section, and the transport path of the target raw material is also located between the two side walls of the sorting section.

[0049] In one embodiment, the sorting section is inclined, and the maximum height of the sorting section in the direction of gravity is less than the height of the transport section in the direction of gravity. The preset obstacle is arranged between the sorting section and the transport section.

[0050] Among them, the fact that the maximum height of the sorting section in the direction of gravity is less than the height of the transport section in the direction of gravity should be understood as: the maximum height of the sorting section in the direction of gravity is less than the minimum height of the transport section in the direction of gravity, or the maximum height of the sorting section in the direction of gravity is less than the height of the end of the transport section in the direction of gravity, where the end of the transport section is the end of the transport section facing the sorting section.

[0051] In this embodiment, the sorting section is inclined, and the maximum height of the sorting section in the direction of gravity is set to be less than the height of the conveying section in the direction of gravity. This can facilitate the transportation of the target raw materials by gravity while helping the target raw materials to spread out better on the sorting section to further inhibit the stacking of the target raw materials. Among them, a preset obstacle is arranged between the sorting section and the conveying section, which can not only collect the collision audio through the preset obstacle, but also use the preset obstacle to exert an external force on the target raw materials to cooperate with the inclined sorting section to better help the target raw materials to spread out.

[0052] It should be noted that the maximum inclination angle of the sorting section is less than the set angle threshold (such as 10 degrees, 15 degrees) to ensure the smooth transportation of the target raw materials on the sorting section and avoid the situation where the target raw materials fly out of the sorting section.

[0053] In some embodiments, the installation position of the preset obstacle can be set independently of the support structure corresponding to the conveyor belt. By setting the independently installed preset obstacle, the vibration noise during the operation of the conveyor belt can be prevented from being mixed into the collision audio, ensuring that the analysis results obtained from the subsequent analysis of the collision audio can maintain a high accuracy. Herein, the support structure corresponding to the conveyor belt can be understood as a rigid structure placed on the ground and used to support the conveyor belt. For example, a support frame formed by combining and splicing several steel materials. For example, the preset obstacle can be installed on the ceiling / roof corresponding to the conveyor belt area, and the preset obstacle extends from the ceiling / roof towards the conveyor belt.

[0054] In some embodiments, there can also be multiple preset obstacles, and the multiple preset obstacles are arranged in an array between the sorting section and the conveying section to collect more comprehensive collision audio for subsequent analysis. For the multiple collision audio collected by the multiple preset obstacles respectively, multiple analysis results can be obtained accordingly. In this case, among the multiple analysis results, the probability data with the highest probability and the corresponding category data are selected to configure the image acquisition frequency and feature library of the target raw materials.

[0055] See Figure 2 , Figure 2 is a schematic structural diagram of an intelligent sorting system 200 for construction waste resource treatment provided by an embodiment of the present invention. As Figure 2 shown, the intelligent sorting system 200 for construction waste resource treatment includes: An audio acquisition module 201, configured to acquire the collision audio of the target raw materials, where the target raw materials are construction waste to be sorted, and the collision audio is the audio generated when the target raw materials collide with a preset obstacle during the transmission process; An audio analysis module 202 is configured to analyze the collision audio to obtain an analysis result. The analysis result includes probability data and category data. The probability data is used to represent the probability that the target raw material includes the target material, and the category data is used to represent the material category of the target material included in the target raw material. The target material is one of a plurality of preset high-value materials. A sorting processing module 203 is configured to control a sorting robot to perform visual sorting processing on the target raw material according to the analysis result. The sorting robot is configured to: collect an image of the target raw material, and when it is detected that the image includes the target material, grab the target material to a target position. The higher the probability represented by the probability data, the higher the image acquisition frequency of the target raw material. The feature library used by the sorting robot during detection is determined based on the category data.

[0056] In one embodiment, the audio analysis module 202 is specifically configured to: Extract features from the collision audio to obtain a first audio feature; Perform similarity calculation between each of a plurality of second audio features and the first audio feature to obtain a plurality of similarity values. The plurality of second audio features correspond to the plurality of high-value materials one by one. The second audio feature is used to represent the characteristics of the audio generated when the corresponding high-value material collides with the preset obstacle. The similarity value is used to represent the feature similarity between the corresponding second audio feature and the first audio feature. Determine the maximum similarity value among the plurality of similarity values as the target similarity value, and generate the analysis result according to the target similarity value and the high-value material corresponding to the target similarity value. The probability data is determined based on the target similarity value, and the category data is determined based on the high-value material corresponding to the target similarity value.

[0057] In one embodiment, the collision audio is obtained based on an audio acquisition device, and the audio sensing end of the audio acquisition device is tightly pressed against the outer wall of the preset obstacle.

[0058] In one embodiment, the target raw material is transported by a conveyor belt, and the forward direction of the target raw material on the conveyor belt is the first direction; The conveyor belt includes a transmission section and a sorting section connected along the first direction. The maximum length of the transmission section in the second direction is less than the minimum length of the sorting section in the second direction, and the length of the sorting section in the second direction gradually increases along the first direction. The second direction is perpendicular to the gravity direction and perpendicular to the first direction. The sorting robot performs visual sorting on the target raw material in the sorting section.

[0059] In one embodiment, the sorting section is inclined, and the maximum height of the sorting section in the gravity direction is less than the height of the transmission section in the gravity direction. The preset obstacle is arranged between the sorting section and the transmission section.

[0060] The intelligent sorting system 200 for construction waste resource treatment can implement Figure 1 each process of the method embodiment in the present invention and achieve the same beneficial effects. To avoid repetition, it will not be elaborated here.

[0061] The embodiment of the present invention also provides an electronic device. Please refer to Figure 3 , the electronic device may include a processor 301, a memory 302, and a program 3021 stored in the memory 302 and executable on the processor 301.

[0062] When the program 3021 is executed by the processor 301, it can implement Figure 1 any step in the corresponding method embodiment and achieve the same beneficial effects. It will not be elaborated here.

[0063] Those of ordinary skill in the art can understand that all or part of the steps of implementing the above method embodiment can be completed by hardware related to program instructions, and the program can be stored in a readable medium.

[0064] The embodiment of the present invention also provides a readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it can implement any step in the above Figure 1 corresponding method embodiment and can achieve the same technical effects. To avoid repetition, it will not be elaborated here.

[0065] The computer-readable storage medium according to an embodiment of the present invention may adopt any combination of one or more computer-readable media. The computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium may be, for example, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples (non-exhaustive list) of the computer-readable storage medium include: an electrical connection having one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this document, the computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0066] The computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries the computer-readable program code. Such a propagated data signal may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer-readable signal medium may also be any computer-readable medium other than the computer-readable storage medium, which can send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0067] The program code contained on the storage medium may be transmitted by any appropriate medium, including but not limited to wireless, wire, optical fiber cable, RF, etc., or any suitable combination of the above.

[0068] The computer program code for performing the operations of the present invention may be written in one or more programming languages or combinations thereof. The programming languages include object-oriented programming languages such as Java, Smalltalk, C++, and also include conventional procedural programming languages such as the "C" language or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, executed as an independent software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or terminal. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (for example, by using an Internet service provider to connect through the Internet).

[0069] The above is the preferred implementation mode of the embodiment of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the principle described in the present invention, several improvements and refinements can still be made, and these improvements and refinements should also be regarded as the protection scope of the present invention.

Claims

1. An intelligent sorting method for resource treatment of construction waste, characterized in that, The method includes: Obtaining the collision audio of the target raw material, where the target raw material is construction waste to be sorted, and the collision audio is the audio generated when the target raw material collides with a preset obstacle during transmission; Analyzing the collision audio to obtain an analysis result, where the analysis result includes probability data and category data. The probability data is used to represent the probability that the target raw material includes the target material, and the category data is used to represent the material category of the target material included in the target raw material. The target material is one of a preset variety of high-value materials; Controlling a sorting robot to perform visual sorting processing on the target raw material according to the analysis result, where the sorting robot is used to: collect an image of the target raw material, and when it is detected that the image includes the target material, grab the target material to a target position; The higher the probability represented by the probability data, the higher the image acquisition frequency of the target raw material, and the feature library used by the sorting robot during detection is determined based on the category data.

2. The method according to claim 1, wherein The analyzing the collision audio to obtain an analysis result includes: Performing feature extraction on the collision audio to obtain a first audio feature; Performing similarity calculation between each second audio feature in a plurality of second audio features and the first audio feature to obtain a plurality of similarity values, where the plurality of second audio features correspond one-to-one to the variety of high-value materials, and the second audio feature is used to represent the feature of the audio generated when the corresponding high-value material collides with the preset obstacle; the similarity value is used to represent the feature similarity between the corresponding second audio feature and the first audio feature; Determining the maximum similarity value among the plurality of similarity values as the target similarity value, and generating the analysis result based on the target similarity value and the high-value material corresponding to the target similarity value, where the probability data is determined based on the target similarity value, and the category data is determined based on the high-value material corresponding to the target similarity value.

3. The method according to claim 1, wherein The collision audio is obtained based on an audio acquisition device, and the audio sensing end of the audio acquisition device is tightly pressed against the outer wall of the preset obstacle.

4. The method according to any one of claims 1-3, characterized in that, The target raw material is transported by a conveyor belt, and the forward direction of the target raw material on the conveyor belt is the first direction; The conveyor belt includes a transmission section and a sorting section connected along the first direction, where the maximum length of the transmission section in the second direction is less than the minimum length of the sorting section in the second direction, and the length of the sorting section in the second direction gradually increases along the first direction. The second direction is perpendicular to the gravity direction and perpendicular to the first direction; The sorting robot performs visual sorting processing on the target raw material in the sorting section.

5. The method according to claim 4, wherein The sorting section is inclined, and the maximum height of the sorting section in the gravity direction is less than the height of the transmission section in the gravity direction. The preset obstacle is arranged between the sorting section and the transmission section.

6. An intelligent sorting system for resource treatment of construction waste, characterized in that, The system includes: An audio acquisition module for acquiring the collision audio of a target raw material, where the target raw material is construction waste to be sorted, and the collision audio is the audio generated when the target raw material collides with a preset obstacle during transmission; An audio analysis module for analyzing the collision audio to obtain an analysis result, where the analysis result includes probability data and category data. The probability data is used to represent the probability that the target raw material includes a target material, and the category data is used to represent the material category of the target material included in the target raw material. The target material is one of a preset variety of high-value materials; A sorting processing module for controlling a sorting robot to perform visual sorting processing on the target raw material according to the analysis result, where the sorting robot is used to: collect an image of the target raw material and, when it detects that the image includes the target material, grab the target material to a target position; The higher the probability represented by the probability data, the higher the image acquisition frequency of the target raw material, and the feature library used by the sorting robot during detection is determined based on the category data.

7. The system according to claim 6, characterized in that, The audio analysis module is specifically used for: extracting features from the collision audio to obtain a first audio feature; performing similarity calculation between each of a plurality of second audio features and the first audio feature to obtain a plurality of similarity values, where the plurality of second audio features correspond one-to-one to the plurality of high-value materials, and the second audio feature is used to represent the characteristics of the audio generated when the corresponding high-value material collides with the preset obstacle; the similarity value is used to represent the feature similarity between the corresponding second audio feature and the first audio feature; determining the maximum similarity value among the plurality of similarity values as the target similarity value, and generating the analysis result based on the target similarity value and the high-value material corresponding to the target similarity value, where the probability data is determined based on the target similarity value, and the category data is determined based on the high-value material corresponding to the target similarity value.

8. The system according to claim 6, wherein The collision audio is obtained based on an audio acquisition device, and the audio sensing end of the audio acquisition device is tightly pressed against the outer wall of the preset obstacle.

9. The system according to any one of claims 6 - 8, characterized in that, The target raw material is transported by a conveyor belt, and the forward direction of the target raw material in the conveyor belt is the first direction; The conveyor belt includes a transmission section and a sorting section connected along the first direction, where the maximum length of the transmission section in the second direction is less than the minimum length of the sorting section in the second direction, and the length of the sorting section in the second direction gradually increases along the first direction. The second direction is perpendicular to the gravity direction and perpendicular to the first direction; The sorting robot performs visual sorting processing on the target raw material in the sorting section.

10. The system according to claim 9, wherein The sorting section is inclined, and the maximum height of the sorting section in the gravity direction is less than the height of the transmission section in the gravity direction. The preset obstacle is arranged between the sorting section and the transmission section.

Images