CN110889305A - Waste sorting apparatus, method, device, calculation device, and storage medium - Google Patents
Waste sorting apparatus, method, device, calculation device, and storage medium Download PDFInfo
- Publication number
- CN110889305A CN110889305A CN201810937181.9A CN201810937181A CN110889305A CN 110889305 A CN110889305 A CN 110889305A CN 201810937181 A CN201810937181 A CN 201810937181A CN 110889305 A CN110889305 A CN 110889305A
- Authority
- CN
- China
- Prior art keywords
- waste
- classification
- unit
- model
- classification model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2413—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on distances to training or reference patterns
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2413—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on distances to training or reference patterns
- G06F18/24147—Distances to closest patterns, e.g. nearest neighbour classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/30—Administration of product recycling or disposal
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02W—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO WASTEWATER TREATMENT OR WASTE MANAGEMENT
- Y02W90/00—Enabling technologies or technologies with a potential or indirect contribution to greenhouse gas [GHG] emissions mitigation
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Business, Economics & Management (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Human Resources & Organizations (AREA)
- Evolutionary Biology (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Sustainable Development (AREA)
- Multimedia (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Image Analysis (AREA)
Abstract
The application discloses waste sorting equipment, a method and a device, a computing device and a storage medium, and relates to the technical field of image processing. In the method, whether a waste product is input is detected; when a waste product is put in, acquiring an image of the waste product; classifying the images based on a waste classification model to obtain a classification result of the waste, wherein the waste classification model is a model which is established in advance according to a deep learning method. The method can automatically detect and classify the waste, and is more accurate and reliable compared with manual classification.
Description
Technical Field
The present application relates to the field of image processing technologies, and in particular, to a waste sorting apparatus, a waste sorting method, a waste sorting apparatus, and a storage medium.
Background
Along with the development of logistics, the development of science and technology and the improvement of the quality of life of people. Human waste products are more and more, and the varieties of the waste products are also more and more. How to treat the waste is a problem which is always concerned.
The key operation of waste treatment in sorting waste products. In the prior art, a throwing person of the waste determines the type of the thrown waste according to own experience, and then the thrown waste is put into a corresponding garbage can. Or after the throwing person of the waste selects the type of the waste through the human-computer interaction interface, throwing the waste into the garbage can corresponding to the pair of types.
However, the classification of the waste is subjectively determined by a human regardless of whether a human-machine interface exists for the delivery of the waste. The classification knowledge of the waste is limited in popularization, people can easily make mistakes, and the garbage can is convenient to artificially discard the waste to the nearest place. Therefore, the classification of the waste products in the prior art is inaccurate and the reliability is low.
Disclosure of Invention
The embodiment of the application provides waste classification equipment, a waste classification method, a waste classification device, a waste classification calculation device and a storage medium, and aims to solve the problems that waste classification is manually operated and accuracy and reliability are low in the prior art.
In a first aspect, the present application provides a waste sorting apparatus, including a waste sensing unit, an image acquisition unit, a data processing unit, a waste moving unit, and a waste storage unit, wherein:
the waste sensing unit is used for detecting whether waste is input;
the image acquisition unit is used for acquiring an image of the waste when the waste sensing unit detects that the waste is put in;
the data processing unit is used for classifying the images based on a waste classification model to obtain a classification result of the waste, wherein the waste classification model is a model which is established in advance according to a deep learning method;
and the waste moving unit is used for moving the waste to the corresponding waste storage unit according to the classification result for storage.
Further, the waste sensing unit is an infrared sensing device.
Further, the device also comprises an output unit and a waste storage amount detection device;
the waste storage amount detection device is arranged in the waste storage unit and used for detecting whether the waste storage unit is full of waste or not and sending information of full waste to the data processing unit when detecting that the waste storage unit is full of waste;
the output unit is communicated with the data processing unit and is used for outputting the information of the full waste according to the control of the data processing unit; the information of the full waste comprises the identification of the waste storage unit of the full waste.
Further, the output unit includes: a voice output unit and/or a display unit.
In a second aspect, the present application provides a method of sorting waste, the method comprising:
detecting whether a waste product is input;
when a waste product is put in, acquiring an image of the waste product;
classifying the images based on a waste classification model to obtain a classification result of the waste, wherein the waste classification model is a model which is established in advance according to a deep learning method.
Further, constructing the waste classification model includes:
acquiring a training sample, wherein the training sample comprises a waste image and a corresponding classification label;
inputting the training samples into a learning model constructed based on a deep learning method to obtain a learning result;
and when the learning result is that the classification accuracy reaches the specified accuracy, obtaining the waste classification model.
Further, the learning model comprises a convolution neural unit, a full connection layer and a classifier which are connected in sequence.
Further, the convolution neural unit comprises three convolution units which are connected in sequence, and each convolution unit comprises a convolution layer, a linear unit and a pooling layer which are connected in sequence.
Further, the classifying the image based on the waste classification model to obtain a classification result of the waste includes:
each convolution unit in the waste classification model adopts a convolution layer thereof to perform convolution operation on the image according to a preset convolution core to obtain a plurality of data sets; obtaining a data set by adopting convolution kernel convolution operation once; then, adjusting the negative numbers in the multiple data sets to be specified values by adopting a linear unit, outputting the data sets to a pooling layer, and pooling the input multiple data sets by the pooling layer to obtain pooled multiple data sets; wherein the specified value is a non-negative number;
splicing the input multiple data sets by adopting a full connection layer in the waste classification model to obtain spliced data;
and classifying operation is carried out on the spliced data by adopting a classifier in the waste classification model, and the classification corresponding to the waste is determined according to the classification operation result.
In a third aspect, the present application provides a waste sorting apparatus comprising:
the detection module is used for detecting whether waste is input;
the image acquisition module is used for acquiring an image of a waste when the waste is thrown in;
and the classification module is used for classifying the images based on a waste classification model to obtain a classification result of the waste, wherein the waste classification model is a model which is established in advance according to a deep learning method.
Further, the apparatus further comprises a model construction module for constructing the waste classification model according to the following method:
acquiring a training sample, wherein the training sample comprises a waste image and a corresponding classification label;
inputting the training samples into a learning model constructed based on a deep learning method to obtain a learning result;
and when the learning result is that the classification accuracy reaches the specified accuracy, obtaining the waste classification model.
Further, the learning model comprises a convolution neural unit, a full connection layer and a classifier which are connected in sequence.
Further, the convolution neural unit comprises three convolution units which are connected in sequence, and each convolution unit comprises a convolution layer, a linear unit and a pooling layer which are connected in sequence.
Further, the classification module comprises:
the convolution pooling unit is used for performing convolution operation on the image by each convolution unit in the waste classification model by adopting a convolution layer of the convolution unit according to a preset convolution core to obtain a plurality of data sets; obtaining a data set by adopting convolution kernel convolution operation once; then, adjusting the negative numbers in the multiple data sets to be specified values by adopting a linear unit, outputting the data sets to a pooling layer, and pooling the input multiple data sets by the pooling layer to obtain pooled multiple data sets; wherein the specified value is a non-negative number;
the full-connection unit is used for splicing a plurality of input data sets by adopting a full-connection layer in the waste classification model to obtain spliced data;
and the classification unit is used for performing classification operation on the spliced data by adopting the classifier in the waste classification model and determining the classification corresponding to the waste according to the classification operation result.
In a fourth aspect, the present application further provides a computing device comprising at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform any of the methods of waste classification provided by embodiments of the present application.
In a fifth aspect, the present application further provides a computer storage medium, wherein the computer storage medium stores computer-executable instructions for causing a computer to perform any one of the methods of classifying waste in the embodiments of the present application.
The waste sorting device, the waste sorting method, the waste sorting device, the waste sorting computing device and the storage medium can automatically detect waste, sort the waste and then place the waste into the corresponding storage unit. People only need to put the waste into the equipment. Thus, the classification of the waste products becomes simple and accurate.
Additional features and advantages of the application will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by the practice of the application. The objectives and other advantages of the application may be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the application and together with the description serve to explain the application and not to limit the application. In the drawings:
fig. 1 is a schematic diagram of an application scenario in an embodiment of the present application;
FIG. 2a is a schematic structural diagram of a waste sorting apparatus according to an embodiment of the present application;
FIG. 2b is a second schematic structural view of a waste sorting apparatus according to an embodiment of the present application;
FIG. 3 is a schematic diagram of an embodiment of the present application showing a waste storage unit full of waste;
FIG. 4 is a schematic flow chart illustrating a waste sorting method according to an embodiment of the present application;
FIG. 5 is a schematic structural diagram of a deep learning mill in an embodiment of the present application;
FIG. 6 is a schematic diagram of a convolutional neural unit in an embodiment of the present application;
FIG. 7 is a schematic structural diagram of a convolution unit in an embodiment of the present application;
FIG. 8 is a schematic structural view of a waste sorting apparatus according to an embodiment of the present application;
fig. 9 is a schematic structural diagram of a computing device according to an embodiment of the present application.
Detailed Description
In order to beautify more naturally, the embodiment of the application provides a beautifying method, a beautifying device, a computing device and a storage medium. In order to better understand the technical solution provided by the embodiments of the present application, the following brief description is made on the basic principle of the solution:
it is difficult to require everyone to have a sophisticated knowledge of garbage classification, especially people who lose garbage, including children and the elderly. Even children who do not know words can lose garbage, and it is difficult to popularize garbage classification for the children. Therefore, the inventors of the present application have studied and developed a scheme for automatically performing classification by a device region. Accordingly, a waste sorting apparatus is proposed. The equipment mainly depends on an artificial intelligence technology to finish garbage classification. The artificial intelligence technology can complete classification functions through the learned knowledge and can continuously accumulate the learned knowledge, so that the artificial intelligence technology can be applied to garbage classification. Specifically, a model established based on a deep learning method may be trained first, so that the model can have sufficient garbage classification knowledge, and the training is finished when the model can accurately perform garbage classification. Then the model can now be applied to the actual garbage classification. In order to simplify the operation of people, the waste sorting device proposed by the inventor can also automatically place waste into the corresponding trash can according to the sorting result.
The waste classification scheme provided by the embodiment of the present application is further described below with reference to the accompanying drawings.
Fig. 1 is a schematic view illustrating a scenario for completing waste classification according to the solution provided in the embodiment of the present application. Included in this scenario are the user's 10 and waste sorting device 11. Further, the scene may further include a server 12, and the server 12 continuously learns and optimizes the waste classification module, and updates the waste classification model in the waste classification device. Of course, the waste classification model may also be stored in the server 12, and the images are collected by the waste classification device and then uploaded to the server 12, so that the server 12 completes the classification.
When the waste classification device finishes classifying, after the user 10 can throw the waste into the waste classification device 11, the waste classification device 11 can sense the waste and take a picture of the waste, the picture is analyzed based on the waste classification model to obtain a classification result of the waste, and then the waste is thrown into a corresponding garbage can.
The waste sorting device 11 and the server 12 may be communicatively connected through a communication network, which may be a local area network, a wide area network, or the like, or may be connected through a line.
The waste sorting apparatus provided in the embodiment of the present application is further described with reference to the accompanying drawings, as shown in fig. 2a, which is a schematic structural diagram of a waste sorting apparatus, the apparatus includes a waste sensing unit 21, an image capturing unit 22, a data processing unit 23, a waste moving unit 24, and a waste storage unit 25, wherein:
the waste sensing unit 21 is used for detecting whether waste is input;
the image acquisition unit 22 is used for acquiring an image of a waste when the waste sensing unit 21 detects that the waste is put in;
the data processing unit 23 is configured to classify the image based on a waste classification model to obtain a classification result of the waste, where the waste classification model is a model that is established in advance according to a deep learning method;
and the waste moving unit 24 is used for moving the waste to the corresponding waste storage unit 25 according to the classification result and storing the waste.
Therefore, the waste sorting device provided by the embodiment of the application can automatically detect the waste, sort the waste and then place the waste into the corresponding storage unit. People only need to put the waste into the equipment. Thus, the classification of the waste products becomes simple and accurate.
Wherein, in one embodiment, any object in nature will always emit infrared radiation outwards as long as the temperature is above absolute zero (-273 ℃). The higher the temperature of the object, the smaller the peak wavelength of the infrared radiation it emits, and the greater the energy of the emitted infrared radiation. When a person enters the sensing range, the infrared sensing device can detect the change of the infrared spectrum of the human body. Therefore, can confirm someone when infrared induction system senses the change of human infrared spectrum in this application embodiment and put in the discarded object, can confirm to detect the discarded object.
Of course, in practice, the sensing of the waste may be determined when the infrared spectrum of the waste is sensed.
Therefore, the waste sensing unit 21 may be an infrared sensing device.
In addition, in this application embodiment, the door that can open and shut can be installed to the abandonment article mouth of puting in among the abandonment article sorting equipment, and people need open this door just can drop into the abandonment article sorting equipment with the abandonment article when puting in the abandonment article. Therefore, in the specific implementation, the detection device can be used for detecting the opening and closing state of the door. When the door is switched from the closed state to the open state, it is determined that a waste product is detected. In summary, any device capable of determining that a waste is input is suitable for the embodiments of the present application, and the present application is not limited thereto.
In one embodiment, the image capturing unit 22 may be any device capable of taking pictures, such as a micro camera, a pinhole camera, and the like, and is suitable for the embodiment of the present application.
Wherein. In one embodiment, the data processing unit 23 may include a processor and a memory, the memory storing data, the processor being used for computational processing. The captured images and the waste classification model may be stored in memory for recall by the processor.
In specific implementation, the processor may adopt an embedded MCU. The purposes of small calculated amount, small occupied system resource and quick real-time response can be achieved. Meanwhile, compared with the PC, the MCU can reduce the cost.
Wherein, in one embodiment, the waste moving unit 24 may be a robot arm. When the classification of the waste is determined, the data processing unit 23 controls the robot arm to clamp the waste into the corresponding waste storage unit.
Furthermore, the waste storage unit can be a garbage can, and the garbage can adopt the existing garbage can with any shape and is suitable for the embodiment of the application.
In the embodiment of the present application, in order to avoid that the waste storage unit can be cleaned in time after being filled with garbage, as shown in fig. 2b, the waste sorting apparatus in the embodiment of the present application further includes an output unit 26 and a waste storage amount detection device 27.
The waste storage amount detection device 27 is installed in the waste storage unit 25, and is used for detecting whether the waste storage unit 25 is full of waste, and sending information of full waste to the data processing unit 23 when detecting full waste;
the output unit 26 is communicated with the data processing unit 23 and is used for outputting the information of the full waste according to the control of the data processing unit 23; the information of the full waste includes the identifier of the waste storage unit 25 in which the full waste is stored.
In a specific implementation, the data processing unit 23 may control the output unit 26 to output the message and the output unit 26 to issue an alarm when receiving the information of the full waste sent by the waste storage amount detection device 27. Thus, the waste disposal personnel can clean the waste disposal personnel in time.
Further, the data processing unit 23 may record the message after receiving the information of the full waste sent from the waste storage amount detection means 27. The type of the waste is determined when a new waste is put into the waste storage device, if the waste storage unit corresponding to the new waste is determined to be full according to the stored information, the control output unit outputs the information of the full waste, and the information can also comprise a classification result, so that waste putting personnel can put the waste into other trash cans needing manual classification again according to the classification result.
Further, the output unit may include a voice output unit and/or a display unit when implemented.
The display unit can be a display screen or a display lamp. As shown in fig. 3, different colored display lights may be included, each of which may be flanked by a corresponding identification of the waste storage unit. Which lamp is on indicates which lamp corresponds to the waste storage unit that is full of waste. As shown in fig. 3, the middle lamp corresponds to a full waste storage unit.
Based on the same inventive concept, an embodiment of the present application further provides a waste classification method, as shown in fig. 4, which is a schematic flow chart of the method, and includes the following steps:
step 401: and detecting whether a waste product is input.
Step 402: when a waste product is put in, an image of the waste product is collected.
In specific implementation, the collected images can be preprocessed and then classified by a waste classification model. The preprocessing may include adjusting the contrast of the image so that the waste in the image can be accurately identified. The preprocessing may also include rotating and cropping the image, and the specific implementation may be determined according to actual requirements, which is not limited in the present application.
Step 403: classifying the images based on a waste classification model to obtain a classification result of the waste, wherein the waste classification model is a model which is established in advance according to a deep learning method.
Wherein, in one embodiment, the waste classification model is constructed according to the following method comprising:
step A1: and acquiring a training sample, wherein the training sample comprises a waste image and a corresponding classification label.
In specific implementation, images of a large number of waste products can be collected, and then waste product classifications corresponding to each image are labeled. In this way, a large number of samples are obtained for model learning. The samples may be divided into training samples and test samples. The training samples are used for training the model, and the testing samples are used for testing the training results of the model.
Step A2: and inputting the training samples into a learning model constructed based on a deep learning method to obtain a learning result.
Step A3: and when the learning result is that the classification accuracy reaches the specified accuracy, obtaining the waste classification model.
Wherein, during training, the learning step length can be set so as to facilitate the training of the model according to the learning step length. And enabling the model to identify the classification result corresponding to the garbage in the picture according to the picture.
The execution accuracy can be set to 98% or more, so that the model can classify the garbage more accurately.
In particular, as shown in fig. 5, the learning model may include a convolutional neural unit 501, a fully-connected layer 502, and a classifier 503 connected in sequence. The convolution nerve unit is used for performing convolution operation on the image. The convolution kernel may be a3 × 3 convolution kernel, but may be another convolution kernel in a specific implementation, which is not limited in the present application. A fully connected layer may integrate class-specific local information of the layer above its connection. In order to improve the performance of the convolutional neural network, a ReLU function can be adopted as the excitation function of each neuron of the full connection layer. The output values of the fully connected layer are passed to a classifier. In particular, the classifier may use softmax logistic regression (softmax regression) for classification. In specific implementation, the model includes a full connection and a classifier.
Further, since too many designs of each layer in the model affect the data processing efficiency, too few designs affect the classification result. In order to select a suitable model, in the embodiment of the present application, as shown in fig. 6, the convolutional neural unit includes three convolutional units (as convolutional units 5011, 5012, and 5013 in fig. 6) connected in sequence, and further, as shown in fig. 7, each convolutional unit includes a convolutional layer 51, a linear unit 52, and a pooling layer 53 connected in sequence. Wherein to reduce overfitting a pooling layer is employed. The adoption of the pooling layer can also reduce the dimension of the data and reduce redundant data so as to facilitate the subsequent accurate classification.
Based on the model designed in fig. 7, in the embodiment of the present application, the classifying the image based on the waste classification model to obtain the classification result of the waste includes:
step A1: each convolution unit in the waste classification model adopts a convolution layer thereof to perform convolution operation on the image according to a preset convolution core to obtain a plurality of data sets; obtaining a data set by adopting convolution kernel convolution operation once; and then, adjusting the negative numbers in the multiple data sets to be designated values by adopting a linear unit, outputting the data sets to a pooling layer, and pooling the input multiple data sets by the pooling layer to obtain pooled multiple data sets.
Wherein the specified value is a non-negative number. In specific implementation, the negative number is converted to 0 by a linear unit, i.e. the specified value in the embodiment of the present application may be 0. Since the pixel value of the image is an RGB value and the value range of the pixel value is between 0 and 255, negative values cannot occur, and the purpose of correcting the processing result can be achieved by adopting a linear unit in the application.
Step A2: and splicing the input data sets by adopting a full connection layer in the waste classification model to obtain spliced data.
Step A3: and classifying operation is carried out on the spliced data by adopting a classifier in the waste classification model, and the classification corresponding to the waste is determined according to the classification operation result.
Further, in this embodiment of the present application, it may also be detected whether the waste storage device is full of waste, and if the waste storage device is full of waste, a message of the full waste storage device is recorded, where the message includes an identifier of the waste storage device full of waste. After the images are classified based on the waste classification model to obtain the classification result of the waste, if the recorded result confirms that the waste storage device corresponding to the waste is full of the waste, the information that the waste storage device is full of the waste is output.
In addition, when it is detected whether the waste storage device is full of waste, a message can be sent so that waste disposal personnel can dispose of the waste in time.
In summary, in the embodiments of the present application, the convolutional neural network is used to classify the waste products without manual classification.
Based on the same inventive concept, the embodiment of the application also provides a waste sorting device. Fig. 8 is a schematic view of the structure of the device. The device includes:
a detection module 801, configured to detect whether a waste is input;
the image acquisition module 802 is used for acquiring an image of a waste when the waste is thrown in;
a classification module 803, configured to classify the image based on a waste classification model to obtain a classification result of the waste, where the waste classification model is a model that is established in advance according to a deep learning method.
Further, the apparatus further comprises a model construction module for constructing the waste classification model according to the following method:
acquiring a training sample, wherein the training sample comprises a waste image and a corresponding classification label;
inputting the training samples into a learning model constructed based on a deep learning method to obtain a learning result;
and when the learning result is that the classification accuracy reaches the specified accuracy, obtaining the waste classification model.
Further, the learning model comprises a convolution neural unit, a full connection layer and a classifier which are connected in sequence.
Further, the convolution neural unit comprises three convolution units which are connected in sequence, and each convolution unit comprises a convolution layer, a linear unit and a pooling layer which are connected in sequence.
Further, the classification module comprises:
the convolution pooling unit is used for performing convolution operation on the image by each convolution unit in the waste classification model by adopting a convolution layer of the convolution unit according to a preset convolution core to obtain a plurality of data sets; obtaining a data set by adopting convolution kernel convolution operation once; then, adjusting the negative numbers in the multiple data sets to be specified values by adopting a linear unit, outputting the data sets to a pooling layer, and pooling the input multiple data sets by the pooling layer to obtain pooled multiple data sets; wherein the specified value is a non-negative number;
the full-connection unit is used for splicing a plurality of input data sets by adopting a full-connection layer in the waste classification model to obtain spliced data;
and the classification unit is used for performing classification operation on the spliced data by adopting the classifier in the waste classification model and determining the classification corresponding to the waste according to the classification operation result.
Having described the beauty method and apparatus of the exemplary embodiments of the present application, a computing apparatus according to another exemplary embodiment of the present application is next described.
As will be appreciated by one skilled in the art, aspects of the present application may be embodied as a system, method or program product. Accordingly, various aspects of the present application may be embodied in the form of: an entirely hardware embodiment, an entirely software embodiment (including firmware, microcode, etc.) or an embodiment combining hardware and software aspects that may all generally be referred to herein as a "circuit," module "or" system.
In some possible embodiments, a computing device according to the present application may include at least one processor, and at least one memory (e.g., the aforementioned first server). The memory stores program codes, and when the program codes are executed by the processor, the processor executes the steps of the system permission opening method according to the various exemplary embodiments of the present application described above in the present specification. For example, the processor may perform steps 401-403 as shown in FIG. 4.
The computing device 130 according to this embodiment of the present application is described below with reference to fig. 9. The computing device 130 shown in fig. 9 is only an example, and should not bring any limitation to the function and the scope of use of the embodiments of the present application.
As shown in fig. 9, computing device 130 is embodied in the form of a general purpose computing device. Components of computing device 130 may include, but are not limited to: the at least one processor 131, the at least one memory 132, and a bus 133 that connects the various system components (including the memory 132 and the processor 131).
The memory 132 may include readable media in the form of volatile memory, such as Random Access Memory (RAM)1321 and/or cache memory 1322, and may further include Read Only Memory (ROM) 1323.
In some possible embodiments, the various aspects of the waste sorting method provided herein may also be implemented in the form of a program product comprising program code for causing a computer device to perform the steps of the beautifying method according to various exemplary embodiments of the present application described above in this specification when the program product is run on the computer device, e.g. the computer device may perform the steps of the beautifying method as shown in fig. 4
Shown as step 401-403.
The program product may employ any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples (a non-exhaustive list) of the readable storage medium include: an electrical connection having one or more wires, a portable 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 disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
The program product for waste sorting of embodiments of the present application may employ a portable compact disc read only memory (CD-ROM) and include program code, and may be run on a computing device. However, the program product of the present application is not limited thereto, and in this document, a readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.
A readable signal medium may include a propagated data signal with readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A readable signal medium may also be any readable medium that is not a readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.
Program code embodied on a readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.
Program code for carrying out operations of the present application may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, C + + or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user computing device, partly on the user equipment, as a stand-alone software package, partly on the user computing device and partly on a remote computing device, or entirely on the remote computing device or server. In the case of remote computing devices, the remote computing device may be connected to the user computing device through any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computing device (e.g., through the internet using an internet service provider).
It should be noted that although several units or sub-units of the apparatus are mentioned in the above detailed description, such division is merely exemplary and not mandatory. Indeed, the features and functions of two or more units described above may be embodied in one unit, according to embodiments of the application. Conversely, the features and functions of one unit described above may be further divided into embodiments by a plurality of units.
Further, while the operations of the methods of the present application are depicted in the drawings in a particular order, this does not require or imply that these operations must be performed in this particular order, or that all of the illustrated operations must be performed, to achieve desirable results. Additionally or alternatively, certain steps may be omitted, multiple steps combined into one step execution, and/or one step broken down into multiple step executions.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
While the preferred embodiments of the present application have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including preferred embodiments and all alterations and modifications as fall within the scope of the application.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present application without departing from the spirit and scope of the application. Thus, if such modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is intended to include such modifications and variations as well.
Claims (12)
1. The utility model provides a waste classification equipment which characterized in that, includes waste induction element, image acquisition unit, data processing unit, waste mobile unit and waste memory cell, wherein:
the waste sensing unit is used for detecting whether waste is input;
the image acquisition unit is used for acquiring an image of the waste when the waste sensing unit detects that the waste is put in;
the data processing unit is used for classifying the images based on a waste classification model to obtain a classification result of the waste, wherein the waste classification model is a model which is established in advance according to a deep learning method;
and the waste moving unit is used for moving the waste to the corresponding waste storage unit according to the classification result for storage.
2. The apparatus of claim 1, wherein the waste sensing unit is an infrared sensing device.
3. The apparatus according to claim 1, further comprising an output unit and a waste storage amount detection device;
the waste storage amount detection device is arranged in the waste storage unit and used for detecting whether the waste storage unit is full of waste or not and sending information of full waste to the data processing unit when detecting that the waste storage unit is full of waste;
the output unit is communicated with the data processing unit and is used for outputting the information of the full waste according to the control of the data processing unit; the information of the full waste comprises the identification of the waste storage unit of the full waste.
4. The apparatus of claim 3, wherein the output unit comprises: a voice output unit and/or a display unit.
5. A method of sorting waste, the method comprising:
detecting whether a waste product is input;
when a waste product is put in, acquiring an image of the waste product;
classifying the images based on a waste classification model to obtain a classification result of the waste, wherein the waste classification model is a model which is established in advance according to a deep learning method.
6. The method of claim 5, wherein constructing the waste classification model comprises:
acquiring a training sample, wherein the training sample comprises a waste image and a corresponding classification label;
inputting the training samples into a learning model constructed based on a deep learning method to obtain a learning result;
and when the learning result is that the classification accuracy reaches the specified accuracy, obtaining the waste classification model.
7. The method of claim 6, wherein the learning model comprises sequentially connected convolutional neural units, fully connected layers, and classifiers.
8. The method of claim 7, wherein the convolutional neural unit comprises three convolutional units connected in sequence, each convolutional unit comprising a convolutional layer, a linear unit, and a pooling layer connected in sequence.
9. The method of claim 8, wherein classifying the image based on a waste classification model to obtain a waste classification result comprises:
each convolution unit in the waste classification model adopts a convolution layer thereof to perform convolution operation on the image according to a preset convolution core to obtain a plurality of data sets; obtaining a data set by adopting convolution kernel convolution operation once; then, adjusting the negative numbers in the multiple data sets to be specified values by adopting a linear unit, outputting the data sets to a pooling layer, and pooling the input multiple data sets by the pooling layer to obtain pooled multiple data sets; wherein the specified value is a non-negative number;
splicing the input multiple data sets by adopting a full connection layer in the waste classification model to obtain spliced data;
and classifying operation is carried out on the spliced data by adopting a classifier in the waste classification model, and the classification corresponding to the waste is determined according to the classification operation result.
10. A waste sorting apparatus, the apparatus comprising:
the detection module is used for detecting whether waste is input;
the image acquisition module is used for acquiring an image of a waste when the waste is thrown in;
and the classification module is used for classifying the images based on a waste classification model to obtain a classification result of the waste, wherein the waste classification model is a model which is established in advance according to a deep learning method.
11. A computer-readable medium having stored thereon computer-executable instructions for performing the method of any one of claims 5-9.
12. An electronic device, comprising:
at least one processor; and
a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,
the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of any one of claims 5-9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810937181.9A CN110889305A (en) | 2018-08-16 | 2018-08-16 | Waste sorting apparatus, method, device, calculation device, and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810937181.9A CN110889305A (en) | 2018-08-16 | 2018-08-16 | Waste sorting apparatus, method, device, calculation device, and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110889305A true CN110889305A (en) | 2020-03-17 |
Family
ID=69744192
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810937181.9A Pending CN110889305A (en) | 2018-08-16 | 2018-08-16 | Waste sorting apparatus, method, device, calculation device, and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110889305A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2006097019A1 (en) * | 2005-03-18 | 2006-09-21 | Xin Wang | A garbage container automatically openable through infrared induction |
| CN205187031U (en) * | 2015-11-20 | 2016-04-27 | 浙江联运知慧科技有限公司 | Intelligence trash classification recycling system |
| CN105772407A (en) * | 2016-01-26 | 2016-07-20 | 耿春茂 | Waste classification robot based on image recognition technology |
| CN205855060U (en) * | 2016-07-04 | 2017-01-04 | 苏州科技大学 | A kind of automatic classification dustbin |
| CN108372126A (en) * | 2017-12-26 | 2018-08-07 | 刘利行 | A kind of Intelligent refuse classification system and its sorting technique |
-
2018
- 2018-08-16 CN CN201810937181.9A patent/CN110889305A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2006097019A1 (en) * | 2005-03-18 | 2006-09-21 | Xin Wang | A garbage container automatically openable through infrared induction |
| CN205187031U (en) * | 2015-11-20 | 2016-04-27 | 浙江联运知慧科技有限公司 | Intelligence trash classification recycling system |
| CN105772407A (en) * | 2016-01-26 | 2016-07-20 | 耿春茂 | Waste classification robot based on image recognition technology |
| CN205855060U (en) * | 2016-07-04 | 2017-01-04 | 苏州科技大学 | A kind of automatic classification dustbin |
| CN108372126A (en) * | 2017-12-26 | 2018-08-07 | 刘利行 | A kind of Intelligent refuse classification system and its sorting technique |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Rabano et al. | Common garbage classification using mobilenet | |
| US20190095764A1 (en) | Method and system for determining objects depicted in images | |
| US20200192927A1 (en) | Detecting affective characteristics of text with gated convolutional encoder-decoder framework | |
| CN110751675B (en) | Urban pet activity track monitoring method based on image recognition and related equipment | |
| CN111401418A (en) | Employee dressing specification detection method based on improved Faster r-cnn | |
| CN106355367A (en) | Warehouse monitoring management device | |
| CN111046956A (en) | Occlusion image detection method and device, electronic equipment and storage medium | |
| Faruk et al. | [Retracted] A Novel and Robust Approach to Detect Tuberculosis Using Transfer Learning | |
| KR20220156603A (en) | Method and electronic device for automated waste management | |
| WO2020056995A1 (en) | Method and device for determining speech fluency degree, computer apparatus, and readable storage medium | |
| US11901066B2 (en) | Express tracking for patient flow management in a distributed environment | |
| US20230048386A1 (en) | Method for detecting defect and method for training model | |
| CN111046849A (en) | Kitchen safety implementation method and device, intelligent terminal and storage medium | |
| CN110276405B (en) | Method and apparatus for outputting information | |
| WO2022121544A1 (en) | Normalizing oct image data | |
| CN113139452A (en) | Method for detecting behavior of using mobile phone based on target detection | |
| Sheiati et al. | Deep learning-based fatigue damage segmentation of wind turbine blades under complex dynamic thermal backgrounds | |
| US20210334590A1 (en) | Automated image curation for machine learning deployments | |
| CN110889305A (en) | Waste sorting apparatus, method, device, calculation device, and storage medium | |
| US10540541B2 (en) | Cognitive image detection and recognition | |
| CN111259698B (en) | Method and device for acquiring image | |
| CN115871679A (en) | Driver fatigue detection method, driver fatigue detection device, electronic device, and medium | |
| KR102342495B1 (en) | Method and Apparatus for Creating Labeling Model with Data Programming | |
| CN114913546A (en) | Method and system for detecting character interaction relationship | |
| Berroukham et al. | Fine-tuning pre-trained vision transformer model for anomaly detection in video sequences |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200317 |
|
| RJ01 | Rejection of invention patent application after publication |