A garbage collection method and apparatus

Abstract

The application provides a garbage recycling method and device, wherein the method comprises the following steps: determining a first pulse number of a spectrum picture captured by a hyperspectral camera; classifying target garbage materials in the spectrum image by using a pre-trained spectrum model to obtain a target image; inputting the target image into a pre-trained neural network model to determine position information of each target garbage in the target image and size information of the target garbage; determining a second pulse number corresponding to each target garbage reaching a preset recycling area according to the position information of each target garbage and the first pulse number; determining a corresponding air jet device of each target garbage according to the size information of the target garbage; for each target garbage, if a current pulse number is greater than or equal to the second pulse number corresponding to the target garbage, then a switch of the air jet device is controlled to be closed, so that the air jet device corresponding to the target garbage sprays gas to blow up the target garbage, so as to recycle the target garbage.

Description

Technical Field

[0001] This application relates to the field of waste treatment technology, and in particular to a waste recycling method and apparatus. Background Technology

[0002] Because there are many types of waste, some can be recycled to reduce environmental pollution. Current technology uses recognition models to identify different types of waste for sorting and recycling. The recognition model needs to quickly identify the location of waste on a high-speed conveyor belt and send this location information to the execution module so that the module can collect the waste. However, due to signal transmission delays between the recognition model and the execution module, the execution module may have difficulty accurately performing the recycling operation based on the location information, resulting in low accuracy in waste recycling. Summary of the Invention

[0003] In view of this, the purpose of this application is to provide at least one waste recycling method and apparatus, which uses a hyperspectral camera to capture spectral images of waste and determine the number of pulses corresponding to the time of the capture. The method identifies the target waste from the spectral images and determines the location and size information of the target waste. It predicts the number of pulses when the target waste arrives at a preset recycling area. When the number of pulses is greater than or equal to the predicted number of pulses, it controls a corresponding jetting device to spray air to blow up the target waste and thus recycle it. This solves the technical problem in the prior art where inaccurate determination of the location information of the target waste leads to inaccurate recycling, and achieves the technical effect of improving the accuracy of determining the target waste.

[0004] This application mainly includes the following aspects:

[0005] In a first aspect, embodiments of this application provide a waste recycling method, the method comprising: determining a first pulse number of a spectral image captured by a hyperspectral camera, the hyperspectral camera being used to capture waste being transported on a waste conveyor belt; classifying the target waste material in the spectral image using a pre-trained spectral model to obtain a target image; inputting the target image into a pre-trained neural network model to determine the position information and size information of each target waste in the target image; determining a second pulse number corresponding to each target waste reaching a preset recycling area based on the position information of each target waste and the first pulse number; determining a jetting device corresponding to each target waste based on the size information of each target waste; for each target waste, determining whether the current pulse number is greater than or equal to the second pulse number corresponding to the target waste; if the current pulse number is greater than or equal to the second pulse number corresponding to the target waste, controlling the jetting switch to close, so that the jetting device corresponding to the target waste ejects gas to blow up the target waste, thereby recycling the target waste.

[0006] Optionally, after determining the first number of pulses received from the hyperspectral camera, the method further includes: determining the third number of pulses received from the hyperspectral camera.

[0007] Optionally, based on the location information of each target waste and the first pulse count, the second pulse count corresponding to the arrival of each target waste in the preset recycling area is determined, including: calculating the pulse count difference between the first pulse count and the third pulse count; determining the pulse increment corresponding to the arrival of each target waste in the preset recycling area based on the location information of each target waste and the movement speed of the waste conveyor belt; and taking the pulse increment corresponding to each target waste, the pulse count difference, and the first pulse count as the second pulse count of each target waste arriving in the preset recycling area.

[0008] Optionally, the location information of each target waste includes: the horizontal coordinate of the target waste in the coordinate system of the waste conveyor belt; determining the pulse increment corresponding to the arrival of each target waste at the preset recycling area based on the location information of each target waste and the movement speed of the waste conveyor belt includes: determining the lateral distance between the horizontal coordinate of each target waste and the horizontal coordinate of the center point of the preset recycling area; determining the transportation time from the center point of the rectangle corresponding to each target waste to the center point of the preset recycling area based on the movement speed and the lateral distance corresponding to each target waste; and using the number of pulses corresponding to the transportation time of each target waste as the pulse increment corresponding to the arrival of each target waste at the preset recycling area.

[0009] Optionally, the location information of each target waste includes: the ordinate of the target waste in the coordinate system of the waste conveyor belt; the size information of each target waste includes: the length of the two mutually perpendicular sides of the rectangle corresponding to the target waste, and the angle between the longer side of the rectangle and the direction of movement of the conveyor belt; determining the jet device corresponding to each target waste based on the size information of each target waste includes: determining the vertical distance between the midpoints of the two shorter sides of the rectangle corresponding to each target waste based on the length of the two mutually perpendicular sides of each target waste and the angle; adding half of the vertical distance corresponding to each target waste to the ordinate of the center point of the target waste as the upper limit value of the ordinate of the target waste; subtracting half of the vertical distance corresponding to the center point of each target waste from the ordinate of the target waste as the lower limit value of the ordinate of the target waste; determining the jet device corresponding to each target waste based on the upper limit value and the lower limit value of the ordinate of the target waste.

[0010] Optionally, after inputting the target image into a pre-trained neural network model to determine the location information and size information of each target piece of waste in the target image, the method further includes: determining whether each target piece of waste is an identified target piece of waste based on the ordinate of each target piece of waste in the waste conveyor belt coordinate system and the size information; if the target piece of waste is an identified target piece of waste, then the target piece of waste is deleted from the target image; if the target piece of waste is not an identified target piece of waste, then the target piece of waste is marked as an identified target piece of waste.

[0011] Optionally, before classifying the target waste material in the spectral image using a pre-trained spectral model to obtain the target image, the method further includes: separating a band image with a preset band range from the spectral image.

[0012] Secondly, embodiments of this application also provide a waste recycling device, comprising: a first determining module, configured to determine the first pulse number of a spectral image received from a hyperspectral camera, the hyperspectral camera being used to photograph waste being transported on a waste conveyor belt; a second determining module, configured to classify the target waste material in the spectral image using a pre-trained spectral model to obtain a target image; a third determining module, configured to input the target image into a pre-trained neural network model to determine the position information and size information of each target waste in the target image; a fourth determining module, configured to determine the second pulse number corresponding to each target waste reaching a preset recycling area based on the position information of each target waste and the first pulse number; a fifth determining module, configured to determine the jetting device corresponding to each target waste based on the size information of each target waste; a comparison module, configured to determine whether the current pulse number is greater than or equal to the second pulse number corresponding to each target waste; and a jetting module, configured to control a jetting switch to close if the current pulse number is greater than or equal to the second pulse number corresponding to the target waste, so that the jetting device corresponding to the target waste ejects gas to blow up the target waste for recycling.

[0013] Thirdly, embodiments of this application also provide an electronic device, including: a processor, a memory, and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor communicates with the memory through the bus, and the machine-readable instructions are executed by the processor to perform the steps of the garbage collection method described in the first aspect or any possible implementation of the first aspect.

[0014] Fourthly, embodiments of this application also provide a computer-readable storage medium storing a computer program, which, when executed by a processor, performs the steps of the garbage collection method described in the first aspect or any possible implementation of the first aspect.

[0015] This application provides a waste recycling method and apparatus. The method includes: determining a first pulse count of a spectral image captured by a hyperspectral camera, the hyperspectral camera being used to capture waste being transported on a waste conveyor belt; classifying the target waste material in the spectral image using a pre-trained spectral model to obtain a target image; inputting the target image into a pre-trained neural network model to determine the position information and size information of each target waste in the target image; determining a second pulse count corresponding to each target waste reaching a preset recycling area based on the position information of each target waste and the first pulse count; determining a jetting device corresponding to each target waste based on the size information of each target waste; for each target waste, determining whether the current pulse count is greater than or equal to the second pulse count corresponding to that target waste; if the current pulse count is greater than or equal to the second pulse count corresponding to that target waste, controlling the jetting switch to close, so that the jetting device corresponding to that target waste ejects gas to blow up the target waste, thereby recycling the target waste. By capturing spectral images of waste using a hyperspectral camera and determining the number of pulses corresponding to the moment of image capture, the target waste is identified from the spectral images, along with its location and size information. The predicted number of pulses when the target waste reaches a preset recycling area is then calculated. When the number of pulses is greater than or equal to the predicted number, the corresponding jetting device is controlled to blow up the target waste, thereby recycling it. This solves the technical problem in existing technologies where inaccurate location information of the target waste leads to inaccurate recycling, thus improving the accuracy of target waste identification.

[0016] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 A flowchart of a waste recycling method provided in an embodiment of this application is shown.

[0019] Figure 2 A schematic diagram of the coordinate system of the waste conveyor belt provided in an embodiment of this application is shown.

[0020] Figure 3 A schematic diagram of the target image provided in an embodiment of this application is shown.

[0021] Figure 4 A functional block diagram of a waste recycling device provided in an embodiment of this application is shown.

[0022] Figure 5 A schematic diagram of the structure of an electronic device provided in an embodiment of this application is shown. Detailed Implementation

[0023] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. It should be understood that the drawings in this application are for illustrative and descriptive purposes only and are not intended to limit the scope of protection of this application. Furthermore, it should be understood that the schematic drawings are not drawn to scale. The flowcharts used in this application illustrate operations implemented according to some embodiments of this application. It should be understood that the operations in the flowcharts may not be implemented in sequence, and steps without logical contextual relationships may be reversed or implemented simultaneously. In addition, those skilled in the art, guided by the content of this application, may add one or more other operations to the flowcharts, or remove one or more operations from the flowcharts.

[0024] Furthermore, the described embodiments are merely some, not all, of the embodiments of this application. The components of the embodiments of this application described and illustrated herein can typically be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of the application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0025] In existing technologies, images of the garbage are taken and their location information at the time of taking the image is determined to calculate the time when the garbage arrives at the preset recycling area. However, due to the high speed of the conveyor belt and the delay in information transmission, the calculated time when the garbage arrives at the preset recycling area is inaccurate, thus affecting the accuracy of garbage recycling.

[0026] Based on this, this application provides a waste recycling method and apparatus. It captures spectral images of waste using a hyperspectral camera and determines the number of pulses corresponding to the capture time. The method identifies the target waste from the spectral images and determines its location and size information. It predicts the number of pulses when the target waste reaches a preset recycling area. When the number of pulses is greater than or equal to the predicted number, it controls a corresponding jetting device to blow away the target waste, thereby recycling it. This solves the technical problem in the prior art where inaccurate determination of the target waste's location information leads to inaccurate recycling, achieving the technical effect of improving the accuracy of target waste identification. Specifically:

[0027] Please see Figure 1 , Figure 1 This is a flowchart illustrating a waste recycling method provided in an embodiment of this application. Figure 1 As shown in the embodiments of this application, the waste recycling method includes the following steps:

[0028] S101: Determine the first pulse number of the received spectral image captured by the hyperspectral camera.

[0029] Hyperspectral cameras are used to photograph garbage being transported on a garbage conveyor belt. The first pulse number refers to the number of pulses corresponding to the received spectral image.

[0030] S102: The target image is obtained by classifying the target waste material in the spectral image using a pre-trained spectral model.

[0031] In other words, a region with the target waste material is cropped from the spectral image using a pre-trained spectral model, and this region becomes the target image. If the pre-trained spectral model fails to crop a region with the target waste material from the spectral image, then the spectral image does not contain waste of the target material, and is therefore deleted, allowing the processing of the next spectral image captured by the hyperspectral camera to continue.

[0032] The spectral image contains multiple pixels, each pixel corresponding to its spectral information, which is the reflectance value of that pixel in each band.

[0033] Specifically, the spectral image is input into a pre-trained spectral model. The spectral model identifies the spectral information of each pixel in the spectral image, thereby finding pixels in the spectral image that have the material of the target waste material. These pixels are then preserved to generate the target image. Consequently, the target image contains only regions with the material of the target waste material.

[0034] The spectral model is trained by using multiple spectral images containing the target waste material as samples and using the pixels corresponding to the target waste material in each spectral image as labels to train the spectral model until the spectral model can remove pixels in the validation set that are not the target waste material, leaving only the pixels corresponding to the target waste material in the images.

[0035] In this application, the target waste material can be a thin film material.

[0036] Before classifying the target waste material in the spectral image using a pre-trained spectral model to obtain the target image, the method further includes: separating a band image with a preset band range from the spectral image.

[0037] In other words, the target image is obtained by first extracting the corresponding band image from the spectral image based on the preset band range, and then classifying the target waste material in the band image using a pre-trained spectral model.

[0038] For example, if the band range of the spectral image is 0μm to 2.35μm, and the preset band range is the near-infrared light band of 0.76μm to 0.90μm, then the image with the band range of 0.76μm to 0.90μm is separated from the spectral image as the band image.

[0039] This application also allows for the installation of cameras at the waste conveyor belt locations, transmitting real-time footage to workers' monitors for real-time monitoring. In case of abnormalities or alarms, the footage can be reviewed. Furthermore, spectral images captured by a hyperspectral camera can be continuously fed into the spectral module for further training, thereby improving the module's accuracy.

[0040] S103: Input the target image into the pre-trained neural network model to determine the location information and size information of each target piece of waste in the target image.

[0041] The garbage with the target garbage material in the spectral image is taken as the target garbage, and thus the target image only contains the target garbage in the spectral image.

[0042] Since the position of the hyperspectral camera is fixed, the location information of each target piece of waste in the spectral image captured by the hyperspectral camera can be obtained.

[0043] Please see Figure 2 , Figure 2 A schematic diagram of the coordinate system for a waste conveyor belt provided in an embodiment of this application is shown. Figure 2As shown, the top view of the garbage conveyor belt can be considered as a rectangle. If the garbage conveyor belt moves from left to right, then the lower left corner of the top view is taken as the origin, the bottom edge of the top view is taken as the horizontal axis, the direction of movement of the garbage conveyor belt is taken as the direction of the horizontal axis, the left vertical edge of the top view is taken as the vertical axis, and the direction vertically upward from the origin is taken as the direction of the vertical axis. If the garbage conveyor belt moves from right to left, then the top view is rotated 180 degrees to the left. The lower left corner of the rotated top view is taken as the origin, the bottom edge of the rotated top view is taken as the horizontal axis, the direction of movement of the garbage conveyor belt is taken as the direction of the horizontal axis (from left to right in the rotated top view), the left vertical edge of the rotated top view is taken as the vertical axis, and the direction vertically upward from the vertical axis is taken as the direction of the vertical axis.

[0044] The location information for each target piece of waste includes its x-coordinate and y-coordinate in the waste conveyor belt coordinate system. The neural network model will mark the bounding box corresponding to the outline of each target piece of waste, and use the x-coordinate of the center point of the bounding box as the x-coordinate of the target piece of waste in the waste conveyor belt coordinate system, and the y-coordinate of the center point of the bounding box as the y-coordinate of the target piece of waste in the waste conveyor belt coordinate system.

[0045] The size information for each target waste includes: the lengths of the two mutually perpendicular sides of the rectangle corresponding to the target waste, and the angle between the long side of the rectangle and the direction of movement of the conveyor belt. In other words, the size information for each target waste includes the lengths of the long side and the short side of the rectangle corresponding to the target waste, and the angle between any long side and the horizontal axis.

[0046] The neural network model is trained as follows: multiple target images are used as samples. For each target image, the labels are: the bounding box corresponding to each manually labeled piece of trash; the lengths of the long and short sides of the bounding box; the angle between any long side of the bounding box and the horizontal axis of the trash conveyor belt coordinate system; and the coordinates of the center point of the bounding box within the trash conveyor belt coordinate system. If the lengths of the long and short sides of the bounding box are the same, then the angle between any long side of the bounding box and the horizontal axis of the trash conveyor belt coordinate system is considered to be 0 degrees.

[0047] S104: Based on the location information of each target waste and the first pulse number, determine the second pulse number corresponding to the arrival of each target waste in the preset recycling area.

[0048] After determining the first number of pulses received from the hyperspectral camera, the method further includes: determining the third number of pulses received from the hyperspectral camera.

[0049] The third pulse number refers to the predicted number of pulses corresponding to the spectral image captured by the hyperspectral camera. Determining the third pulse number of the spectral image captured by the hyperspectral camera includes: inputting the first pulse number into a trained compensation model to obtain the predicted third pulse number; or, subtracting a manually set pulse number difference from the first pulse number as the third pulse number.

[0050] The compensation model can be trained in the following way: obtain the real time corresponding to each shot taken by the hyperspectral camera, and the user converts the real time corresponding to each shot into the third pulse number; use the first pulse number of the received spectral image for each shot as sample data, and use the third pulse number corresponding to each shot as the label to train the compensation model.

[0051] Based on the location information of each target waste and the first pulse count, the second pulse count corresponding to the arrival of each target waste in the preset recycling area is determined, including: calculating the pulse count difference between the first pulse count and the third pulse count; based on the location information of each target waste and the movement speed of the waste conveyor belt, determining the pulse increment corresponding to the arrival of each target waste in the preset recycling area; and taking the pulse increment corresponding to each target waste, the pulse count difference, and the first pulse count as the second pulse count of each target waste arriving in the preset recycling area.

[0052] The step of determining the pulse increment corresponding to the arrival of each target waste in the preset recycling area based on the location information of each target waste and the movement speed of the waste conveyor belt includes: determining the lateral distance between the horizontal coordinate of each target waste and the horizontal coordinate of the center point of the preset recycling area; determining the transportation time from the center point of the rectangle corresponding to each target waste to the center point of the preset recycling area based on the movement speed and the lateral distance corresponding to each target waste; and using the number of pulses corresponding to the transportation time of each target waste as the pulse increment corresponding to the arrival of each target waste in the preset recycling area.

[0053] In other words, for each target waste, the abscissa of the center point of the preset recycling area in the waste conveyor belt coordinate system is subtracted from the abscissa of the target waste, and the difference is taken as the lateral distance of the target waste; the lateral distance of the target waste is compared with the movement speed of the waste conveyor belt, and the ratio is taken as the transportation time for the center point of the rectangle corresponding to the target waste to move to the center point of the preset recycling area, or in other words, the ratio is taken as the transportation time required for the target waste to move to the preset recycling area; the movement time of the target waste is compared with the pulse period, and the ratio is taken as the number of pulses corresponding to the transportation time of the target waste, and the number of pulses is taken as the pulse increment corresponding to the target waste reaching the preset recycling area.

[0054] In this application, the pulses are sent at regular intervals, and the pulse period remains constant.

[0055] In other words, for each target waste, the difference between the first pulse count and the third pulse count of that target waste is taken as the pulse count difference of that target waste. The sum of the pulse increment corresponding to each target waste, the pulse count difference of that target waste, and the first pulse count of that target waste is taken as the predicted second pulse count for each target waste to reach the preset recycling area.

[0056] S105: Based on the size information of each target waste, determine the corresponding jetting device for each target waste.

[0057] Please see Figure 3 , Figure 3 A schematic diagram of the target image provided in an embodiment of this application is shown. Figure 3 As shown, the target image contains three pieces of trash. The location information of the first piece of trash is (x1, y1), and its size information is (length of the long side is a1, length of the short side is b1, and the included angle is ϴ1); the location information of the second piece of trash is (x2, y2), and its size information is (length of the long side is a2, length of the short side is b2, and the included angle is ϴ2); the location information of the third piece of trash is (x3, y3), and its size information is (length of the long side and the short side are always c, and the included angle is 0).

[0058] The speed of the waste conveyor belt is typically 3 to 6 meters per second. The size information of each target waste includes: the lengths of the two mutually perpendicular sides of the rectangle corresponding to the target waste, and the angle between the longer side of the rectangle and the direction of movement of the conveyor belt.

[0059] The step of determining the jetting device corresponding to each target waste based on its size information includes: determining the vertical distance between the midpoints of the two shorter sides of the rectangle corresponding to each target waste based on the lengths of the two mutually perpendicular sides of each target waste and the included angle; adding half of the vertical distance corresponding to each target waste to the ordinate of the center point of the target waste to obtain the upper limit value of the ordinate of the target waste; subtracting half of the vertical distance corresponding to the center point of each target waste from the ordinate of the target waste to obtain the lower limit value of the ordinate of the target waste; and determining the jetting device corresponding to each target waste based on the upper limit value and the lower limit value of the ordinate of the target waste.

[0060] For example, such as Figure 3 As shown, the vertical distance h1 of the first target trash is calculated as follows: The lower limit of the ordinate of the first target trash is y1 minus 0.5×h1, and the upper limit of the ordinate of the first target trash is y1 plus 0.5×h1; the vertical distance h2 of the second target trash is calculated as follows: The lower limit of the vertical coordinate of the second target trash is y2 minus 0.5×h2, and the upper limit of the vertical coordinate of the second target trash is y2 plus 0.5×h2; the vertical distance of the third target trash is the side length c of the rectangle, the lower limit of the vertical coordinate of the third target trash is y3 minus 0.5×c, and the upper limit of the vertical coordinate of the second target trash is y3 plus 0.5×c.

[0061] Furthermore, among all the jetting devices, those with a ordinate within the range of (y1-0.5×h1, y1+0.5×h1) are selected as the jetting devices corresponding to the first target waste; among all the jetting devices, those with a ordinate within the range of (y2-0.5×h2, y2+0.5×h2) are selected as the jetting devices corresponding to the second target waste; and among all the jetting devices, those with a ordinate within the range of (y3-0.5×c, y3+0.5×c) are selected as the jetting devices corresponding to the third target waste.

[0062] After inputting the target image into a pre-trained neural network model to determine the location information and size information of each target piece of waste in the target image, the method further includes: determining whether each target piece of waste is an identified target piece of waste based on the ordinate of each target piece of waste in the waste conveyor belt coordinate system and the size information; if the target piece of waste is an identified target piece of waste, then the target piece of waste is deleted from the target image; if the target piece of waste is not an identified target piece of waste, then the target piece of waste is marked as an identified target piece of waste.

[0063] Because hyperspectral cameras capture a large area, they inevitably capture duplicate images of the same target debris in each spectral image. Therefore, target debris already captured in previous target images is identified as already marked. In the current target image, marked target debris is removed, and unmarked target debris is identified. This prevents the error of repeatedly spraying the same target debris when it is subsequently blown up by the jet device.

[0064] S106: For each target garbage, determine whether the current pulse number is greater than or equal to the second pulse number corresponding to that target garbage.

[0065] In other words, it determines in real time whether the current pulse count is greater than or equal to the second pulse count corresponding to each target piece of waste; if the current pulse count is less than the second pulse count corresponding to the target piece of waste, the jet switch is not controlled to close.

[0066] This application employs a distributed execution approach to determine whether the current pulse count is greater than or equal to the second pulse count corresponding to each target garbage. In other words, each target garbage is processed by a dedicated core processor, and each core processor determines whether the current pulse count is greater than or equal to the second pulse count of the target garbage corresponding to that core processor. This improves processing speed, prevents mutual interference, ensures stable and continuous system operation, and enhances the flexibility and convenience of system maintenance.

[0067] S107: Control the jet switch to close so that the jet device corresponding to the target waste sprays gas to blow up the target waste, so as to recycle the target waste.

[0068] If the current pulse count is greater than or equal to the second pulse count corresponding to the target waste, the jet switch is closed to cause the jet device corresponding to the target waste to spray gas to blow up the target waste and recycle it.

[0069] If the current pulse count is greater than or equal to the second pulse count corresponding to the target waste, the core processor corresponding to the target waste will be interrupted, controlling the jet on / off switch and starting a timer. If the timer duration is greater than or equal to a preset jet duration, the jet switch will be opened, thus stopping the jetting device corresponding to the target waste from ejecting gas. This application uses pulse count instead of the core processor's own time for real-time calculation, avoiding millisecond-level switching losses during core processor switching and preventing impact on tracking accuracy. Furthermore, by using a core processor interrupt-based processing method, the switching time of the core processor is completely avoided, and the horizontal axis coordinate range of the target waste can be shared across the caches of multiple core processors.

[0070] Based on the same application concept, this application also provides a waste recycling device corresponding to the waste recycling method provided in the above embodiments. Since the principle of the device in this application is similar to the waste recycling method in the above embodiments of this application, the implementation of the device can refer to the implementation of the method, and the repeated parts will not be described again.

[0071] like Figure 4 As shown, Figure 4 This is a functional block diagram of a waste recycling device provided in an embodiment of this application. The waste recycling device 10 includes: a first determining module 101, a second determining module 102, a third determining module 103, a fourth determining module 104, a fifth determining module 105, a comparison module 106, and an air jet module 107.

[0072] The first determining module 101 is used to determine the first pulse number of the received spectral image captured by the hyperspectral camera, which is used to capture images of garbage being transported on a garbage conveyor belt.

[0073] The second determining module 102 is used to classify the target waste material in the spectral image using a pre-trained spectral model to obtain the target image.

[0074] The third determining module 103 is used to input the target image into a pre-trained neural network model to determine the location information and size information of each target waste in the target image;

[0075] The fourth determining module 104 is used to determine the second pulse number corresponding to each target waste reaching the preset recycling area based on the location information of each target waste and the first pulse number;

[0076] The fifth determining module 105 is used to determine the jetting device corresponding to each target waste based on the size information of each target waste.

[0077] Comparison module 106 is used to determine, for each target garbage, whether the current pulse number is greater than or equal to the second pulse number corresponding to that target garbage;

[0078] The jet module 107 is used to control the jet switch to close if the current pulse number is greater than or equal to the second pulse number corresponding to the target waste, so that the jet device corresponding to the target waste sprays gas to blow up the target waste and recycle the target waste.

[0079] Based on the same application concept, see [link / reference] Figure 5 The diagram shown is a structural schematic of an electronic device provided in an embodiment of this application. The electronic device 20 includes a processor 201, a memory 202, and a bus 203. The memory 202 stores machine-readable instructions that can be executed by the processor 201. When the electronic device 20 is running, the processor 201 and the memory 202 communicate through the bus 203. The machine-readable instructions are executed by the processor 201 to perform the steps of any of the garbage collection methods described in the above embodiments.

[0080] Specifically, when the machine-readable instructions are executed by the processor 201, they can perform the following processing: determining the first pulse number of a spectral image captured by a hyperspectral camera, the hyperspectral camera being used to capture images of garbage being transported on a garbage conveyor belt; classifying the target garbage material in the spectral image using a pre-trained spectral model to obtain a target image; inputting the target image into a pre-trained neural network model to determine the position information and size information of each target garbage in the target image; determining the second pulse number corresponding to each target garbage reaching a preset recycling area based on the position information of each target garbage and the first pulse number; determining the jetting device corresponding to each target garbage based on the size information of each target garbage; for each target garbage, determining whether the current pulse number is greater than or equal to the second pulse number corresponding to that target garbage; if the current pulse number is greater than or equal to the second pulse number corresponding to that target garbage, controlling the jetting switch to close, so that the jetting device corresponding to that target garbage ejects gas to blow up the target garbage, thereby recycling the target garbage.

[0081] Based on the same concept, embodiments of this application also provide a computer-readable storage medium storing a computer program, which, when run by a processor, performs the steps of the garbage collection method provided in the above embodiments.

[0082] Specifically, the storage medium can be a general-purpose storage medium, such as a portable disk or hard disk. When the computer program on the storage medium is run, it can execute the above-mentioned waste recycling method. It takes a spectral image of the waste using a hyperspectral camera and determines the number of pulses corresponding to the time of the image capture. It identifies the target waste from the spectral image and determines the location and size information of the target waste. It predicts the number of pulses when the target waste reaches the preset recycling area. When the number of pulses is greater than or equal to the predicted number of pulses, it controls the corresponding jet device to spray air to blow up the target waste and thus recycle it. This solves the technical problem in the prior art where inaccurate determination of the location information of the target waste leads to inaccurate recycling, and achieves the technical effect of improving the accuracy of determining the target waste.

[0083] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems and devices described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here. In the several embodiments provided in this application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division; in actual implementation, there may be other division methods. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Another point is that the displayed or discussed mutual coupling or direct coupling or communication connection may be through some communication interfaces; the indirect coupling or communication connection of devices or units may be electrical, mechanical, or other forms.

[0084] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0085] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0086] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a processor-executable, non-volatile, computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0087] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A waste recycling method, characterized in that, The method includes: Determine the first pulse number of the spectral image received from the hyperspectral camera, which is used to photograph garbage being transported on a garbage conveyor belt; The target image is obtained by classifying the target waste material in the spectral image using a pre-trained spectral model. The target image is input into a pre-trained neural network model to determine the location information and size information of each target piece of waste in the target image. Based on the location information of each target waste and the first pulse count, determine the second pulse count corresponding to the arrival of each target waste in the preset recycling area; Based on the size information of each target piece of waste, determine the corresponding jetting device for each target piece of waste; For each target piece of trash, determine whether the current pulse count is greater than or equal to the second pulse count corresponding to that target piece of trash; If the current pulse count is greater than or equal to the second pulse count corresponding to the target waste, the jet switch is closed so that the jet device corresponding to the target waste sprays gas to blow up the target waste and recycle it. After determining the first number of pulses in the received spectral image captured by the hyperspectral camera, the method further includes: determining the third number of pulses in the received spectral image captured by the hyperspectral camera; Determining the third pulse number of the spectral image captured by the hyperspectral camera includes: inputting the first pulse number into the trained compensation model to obtain the predicted third pulse number; The step of determining the second pulse number corresponding to each target waste reaching the preset recycling area based on the location information of each target waste and the first pulse number includes: Calculate the pulse number difference between the first pulse number and the third pulse number; Based on the location information of each target waste and the speed of the waste conveyor belt, determine the number of pulse increments corresponding to each target waste reaching the preset recycling area; The pulse increment for each target waste, the pulse difference, and the sum of the first pulse count are used as the second pulse count for each target waste to reach the preset recycling area.

2. The method according to claim 1, characterized in that, The location information of each target waste includes: the horizontal coordinate of the target waste in the coordinate system of the waste conveyor belt; the determination of the pulse increment number corresponding to the arrival of each target waste in the preset recycling area based on the location information of each target waste and the movement speed of the waste conveyor belt includes: Determine the lateral distance between the x-coordinate of each target waste and the x-coordinate of the center point of the preset recycling area; Based on the movement speed and the lateral distance corresponding to each target waste, the transportation time from the center point of the rectangle corresponding to each target waste to the center point of the preset recycling area is determined; The number of pulses corresponding to the transportation time of each target waste is used as the pulse increment number corresponding to the arrival of each target waste in the preset recycling area.

3. The method according to claim 1, characterized in that, The location information of each target waste includes: the ordinate of the target waste in the coordinate system of the waste conveyor belt; the size information of each target waste includes: the length of the two mutually perpendicular sides of the rectangle corresponding to the target waste, and the angle between the long side of the rectangle and the direction of movement of the waste conveyor belt. The step of determining the corresponding jetting device for each target waste based on its size information includes: Based on the lengths of the two mutually perpendicular sides of each target waste and the included angle, determine the vertical distance between the midpoints of the two short sides of the rectangle corresponding to each target waste. Add half of the vertical distance corresponding to each target piece of waste to the ordinate of the center point of that target piece of waste, and use this as the upper limit value of the ordinate of that target piece of waste; Subtract half of the vertical distance corresponding to the center point of each target waste from the ordinate of the target waste to obtain the lower limit value of the ordinate of the target waste; Based on the upper limit and lower limit of the vertical coordinate corresponding to each target piece of waste, the corresponding jetting device is determined.

4. The method according to claim 3, characterized in that, After inputting the target image into a pre-trained neural network model to determine the location information and size information of each piece of trash in the target image, the method further includes: Based on the ordinate of each target waste in the target image in the coordinate system of the waste conveyor belt and the size information, it is determined whether each target waste is an identified target waste; If the target trash is an identified target trash, then the target trash is removed from the target image; If the target trash is not an identified target trash, then mark the target trash as an identified target trash.

5. The method according to claim 1, characterized in that, Before classifying the target waste material in the spectral image using a pre-trained spectral model to obtain the target image, the method further includes: Separate the band image within the preset band range from the spectral image.

6. A waste recycling device, characterized in that, The apparatus for performing the waste recycling method as described in any one of claims 1 to 5 includes: The first determining module is used to determine the first pulse number of the received spectral image captured by the hyperspectral camera, which is used to capture images of garbage being transported on a garbage conveyor belt. The second determining module is used to classify the target waste material in the spectral image using a pre-trained spectral model to obtain the target image. The third determining module is used to input the target image into a pre-trained neural network model to determine the location information and size information of each target piece of waste in the target image. The fourth determining module is used to determine the second pulse number corresponding to each target waste reaching the preset recycling area based on the location information of each target waste and the first pulse number; The fifth determining module is used to determine the corresponding jetting device for each target waste based on the size information of each target waste. The comparison module is used to determine, for each target garbage, whether the current pulse number is greater than or equal to the second pulse number corresponding to that target garbage; The jet module is used to control the jet switch to close if the current pulse number is greater than or equal to the second pulse number corresponding to the target waste, so that the jet device corresponding to the target waste sprays gas to blow up the target waste and recycle the target waste.

7. An electronic device, characterized in that, include: The device includes a processor, a memory, and a bus. The memory stores machine-readable instructions executable by the processor. When the electronic device is running, the processor communicates with the memory via the bus. The machine-readable instructions are executed by the processor to perform the steps of the garbage collection method as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, performs the steps of the garbage collection method as described in any one of claims 1 to 5.

Images