CN113135368B - Intelligent garbage front-end classification system and method - Google Patents
Intelligent garbage front-end classification system and method Download PDFInfo
- Publication number
- CN113135368B CN113135368B CN202110475037.XA CN202110475037A CN113135368B CN 113135368 B CN113135368 B CN 113135368B CN 202110475037 A CN202110475037 A CN 202110475037A CN 113135368 B CN113135368 B CN 113135368B
- Authority
- CN
- China
- Prior art keywords
- garbage
- image
- module
- information
- classification
- 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.)
- Active
Links
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B65—CONVEYING; PACKING; STORING; HANDLING THIN OR FILAMENTARY MATERIAL
- B65F—GATHERING OR REMOVAL OF DOMESTIC OR LIKE REFUSE
- B65F1/00—Refuse receptacles; Accessories therefor
- B65F1/0033—Refuse receptacles; Accessories therefor specially adapted for segregated refuse collecting, e.g. receptacles with several compartments; Combination of receptacles
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B65—CONVEYING; PACKING; STORING; HANDLING THIN OR FILAMENTARY MATERIAL
- B65F—GATHERING OR REMOVAL OF DOMESTIC OR LIKE REFUSE
- B65F1/00—Refuse receptacles; Accessories therefor
- B65F1/14—Other constructional features; Accessories
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B65—CONVEYING; PACKING; STORING; HANDLING THIN OR FILAMENTARY MATERIAL
- B65F—GATHERING OR REMOVAL OF DOMESTIC OR LIKE REFUSE
- B65F1/00—Refuse receptacles; Accessories therefor
- B65F1/0033—Refuse receptacles; Accessories therefor specially adapted for segregated refuse collecting, e.g. receptacles with several compartments; Combination of receptacles
- B65F2001/008—Means for automatically selecting the receptacle in which refuse should be placed
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B65—CONVEYING; PACKING; STORING; HANDLING THIN OR FILAMENTARY MATERIAL
- B65F—GATHERING OR REMOVAL OF DOMESTIC OR LIKE REFUSE
- B65F2210/00—Equipment of refuse receptacles
- B65F2210/124—Counting means
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B65—CONVEYING; PACKING; STORING; HANDLING THIN OR FILAMENTARY MATERIAL
- B65F—GATHERING OR REMOVAL OF DOMESTIC OR LIKE REFUSE
- B65F2210/00—Equipment of refuse receptacles
- B65F2210/138—Identification means
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B65—CONVEYING; PACKING; STORING; HANDLING THIN OR FILAMENTARY MATERIAL
- B65F—GATHERING OR REMOVAL OF DOMESTIC OR LIKE REFUSE
- B65F2210/00—Equipment of refuse receptacles
- B65F2210/196—Tape dispensers
Landscapes
- Engineering & Computer Science (AREA)
- Mechanical Engineering (AREA)
- Image Processing (AREA)
- Sorting Of Articles (AREA)
Abstract
The invention provides an intelligent garbage front-end classification system and a method, wherein the intelligent garbage front-end classification system comprises: the system comprises a garbage image recognition system, an automatic garbage classification and delivery system and a garbage information processing system. In the application, based on a garbage image recognition system, a garbage automatic classification and delivery system and a garbage information processing system, the intelligent garbage front-end classification system can independently run in a distributed mode, automatically recognize and automatically classify delivered garbage, can collect and feed back garbage classification information, and can update extended functions in a networking mode.
Description
Technical Field
The invention relates to the field of garbage classification treatment, in particular to an intelligent garbage front-end classification system and method.
Background
Due to the requirements of environmental protection and resource recovery, the nation puts forward strict garbage classification management requirements, and the current garbage classification standard is to classify garbage into four categories: can recover garbage, kitchen garbage, harmful garbage and other garbage. The error rate is high due to the limitation of the knowledge structure of the artificial garbage classification; the mechanical or electronic induction garbage classification mainly utilizes material physical properties or basic shapes and colors to classify garbage, is only suitable for specific industrial fields such as metal garbage, building garbage and the like, and has poor expandability. The garbage front-end classification mainly treats daily household garbage, has no fixed shape and physical property, is various in types and complex in background, and is difficult to effectively classify by adopting a traditional method. Meanwhile, the existing garbage front-end classification system is mainly controlled by a remote server or a workstation, and the application occasions of garbage front-end classification are dispersive and high in real-time performance, so that the existing garbage front-end classification system cannot independently, quickly and accurately respond to garbage classification work.
Disclosure of Invention
The invention aims to: in order to solve the technical problems existing in the prior art, the intelligent garbage front-end classification system and method which can operate independently and in a distributed mode, automatically identify, automatically classify and release and advertise and guide are realized.
The technical purpose is realized by the following technical method: an intelligent garbage front-end classification system, comprising: the system comprises a garbage image recognition system, an automatic garbage classification and delivery system and a garbage information processing system.
The garbage image recognition system is an embedded image recognition system, is deployed on a garbage front-end classification device, and is used for image acquisition and recognition of garbage. And the image acquisition of the garbage is realized by carrying out visual acquisition through a camera to obtain the image of the garbage. And the image recognition of the garbage is realized by loading the data of the pre-training model and comparing the data with the collected image to recognize the specific name or the belonging type of the garbage.
The automatic garbage sorting and throwing system comprises an image recognition system, a garbage putting device, a throwing mechanism and a sorting garbage bin, wherein the image recognition system recognizes the specific name or the type of garbage, and the throwing mechanism automatically throws the garbage into the sorting garbage bin corresponding to the garbage.
The garbage information processing system is an information data acquisition and feedback system. The garbage information processing system detects the garbage classification working condition and feeds back information in real time, collects and feeds back the type and the quantity corresponding to the thrown garbage, detects the full load of the classification garbage bin and feeds back the information, and plays the garbage classification promo piece in a circulating mode.
An intelligent garbage front-end classification method is based on the intelligent garbage front-end classification system and comprises the following steps: the system comprises a garbage image recognition system, an automatic garbage classification and delivery system and a garbage information processing system.
Starting the intelligent garbage front-end classification system, starting system operation timing, carrying out visual acquisition by the camera, acquiring a frame of image by the video acquired by the visual acquisition for preprocessing and decoding, carrying out initial calibration on the image, and loading the pre-training model.
In the standby process, a display screen of the garbage information processing system circularly plays garbage classification promos, and propagates garbage classification knowledge and a using method of the intelligent garbage front-end classification system.
In the standby process, the camera carries out visual acquisition and object detection. And acquiring a frame of image for decoding, comparing the frame of image with the initial calibration, and judging whether garbage is put in.
When garbage is put into the object placing device, the garbage image recognition system acquires a frame of image for preprocessing and decoding, compares the frame of image with the initial calibration image, and judges that the object placing device is placed with garbage. And further extracting high-order abstract features and data of the pre-training model for detection and identification, and acquiring probability values of all label values, wherein if the probability value is the maximum or exceeds a threshold value, the label value corresponding to the probability value is assigned to the garbage, and the garbage marks specific garbage name information corresponding to the garbage. And classifying according to the specific garbage name information to obtain the category of the garbage. The types of the garbage are four, and the garbage, kitchen garbage, harmful garbage and other garbage can be recycled.
According to the classification of the garbage, the throwing mechanism automatically throws the garbage into the classification garbage boxes corresponding to the garbage classification, wherein the classification garbage boxes comprise harmful garbage boxes, recyclable garbage boxes, kitchen garbage boxes and other garbage boxes.
And in the working process of the throwing mechanism, the garbage information processing system counts the garbage throwing and updates the garbage quantity information of the corresponding garbage can.
In the working process of the throwing mechanism, the garbage information processing system carries out full load detection on the garbage can to obtain full load information of the garbage can.
And the full load detection is based on the upper limit of the capacity of the dustbin, the measurement of the upper limit of the capacity of the dustbin is based on the height position of the rubbish of the dustbin, the height position of the rubbish is measured, and the full load information of the dustbin is fed back.
The garbage classification propaganda film is played in a stop-loop mode through the display screen, the video acquired by the camera vision is displayed in real time, the process of garbage automatic identification and automatic classification is displayed in real time, the specific garbage name information is displayed in real time, the classification information of the garbage is displayed in real time, the counting number is displayed in real time, the garbage quantity information is displayed in real time, and the full load information of the garbage can is displayed in real time.
And when the garbage can is fully loaded, the garbage information processing system sends information to a remote server through a network to clean the classified garbage can.
The intelligent garbage front-end classification system is connected with a remote server end through a network to update and upgrade so as to realize the updating of garbage identification types and the expansion of garbage identification quantity.
Compared with the prior art, the invention has the following beneficial effects: the intelligent garbage front-end classification system is deployed in a garbage front-end classification processing device, can independently run without a remote server or a workstation, and is strong in instantaneity, high in classification precision and high in stability. The garbage classification is carried out in an image recognition mode, is not limited by the variety, material and form of garbage, and is suitable for the diversity and complexity of daily household garbage. And automatically classifying the garbage through a garbage image recognition system, assigning specific name information of the garbage and assigning garbage category information. The pre-training model can be downloaded and updated through a network and deployed to a garbage front-end classification processing device, and the method is beneficial to improving the expansibility and the precision of garbage recognition. The automatic garbage sorting and throwing system automatically throws garbage into the corresponding sorting garbage bin. The garbage information processing system can carry out garbage input counting, can carry out full-load detection of the garbage can, is full-load and can send information to a remote server end through a network, and is beneficial to the cleaning efficiency and the cost management of the garbage can. The garbage information processing system plays the garbage classification promo sheet when no garbage is thrown in, and feeds back the specific name information and the classification information of the garbage in real time when the garbage is thrown in, so that the garbage classification awareness of a user is guided and enhanced.
Drawings
To further illustrate the invention, the following brief description of the embodiments or drawings required in the description of the prior art is not to be construed as limiting the invention.
Fig. 1 is a schematic diagram of a framework of the intelligent garbage front-end classification system of the present invention.
Fig. 2 is a schematic diagram of a framework of the spam image recognition system of the present invention.
Fig. 3 is a schematic diagram of a framework of the automatic garbage sorting and throwing system of the present invention.
FIG. 4 is a block diagram of a spam processing system in accordance with the present invention.
Fig. 5 is a schematic view of the garbage image recognition process of the present invention.
Fig. 6 is a schematic diagram of the automatic garbage sorting and throwing process of the present invention.
FIG. 7 is a flow chart illustrating garbage information processing according to the present invention.
Detailed Description
The following specific examples are to be construed as merely illustrative, and not limitative of the remainder of the disclosure in any way whatsoever, and after reading this disclosure, persons skilled in the art will understand that various modifications can be made without inventive faculty, and such modifications are to be included within the scope of the appended claims.
The present invention will be described in detail below by way of examples with reference to the accompanying drawings.
An intelligent garbage front-end classification system, comprising: the system comprises a garbage image recognition system 1, an automatic garbage classification and delivery system 2 and a garbage information processing system 3.
The garbage image recognition system 1 comprises an image acquisition module 11, an image feature processing module 12 and an image recognition module 13.
The image recognition module 13 includes an image feature recognition comparison unit 131 and a pre-training model loading unit 132.
The automatic garbage sorting and throwing system 2 comprises a throwing mechanism 21, a storage device 22 and a sorting garbage can 23.
The sorting garbage box 23 includes a recyclable garbage box 231, a kitchen garbage box 232, a harmful garbage box 233 and other garbage boxes 234.
The junk information processing system 3 includes a junk counting module 31, a full load detection module 32, a junk information recording module 33, an information display module 34, an information voice module 35, and a communication transmission module 36.
The image acquisition module 11 includes a camera 111, a light supplement device 112, and a light shielding plate or a light shielding cover 113, where the camera 111 is configured to capture a real-time image, and the light supplement device 112 and the light shielding plate or the light shielding cover 113 are configured to reduce an influence of strong ambient light from the outside and eliminate a shadow.
The throwing mechanism 21 comprises a bottom layer motor or a steering engine 211 and an upper layer motor or a steering engine 212, and the angle of the object placing device 22 is changed or/and the phase of the classification garbage can 23 is changed through the combined motion of the bottom layer motor or the steering engine 211 and the upper layer motor or the steering engine 212, so that garbage is thrown in a classification mode.
If the upper layer motor or the steering engine 212 executes a predetermined action, the feeding mechanism 21 triggers a counting signal, the garbage counting module 31 records the garbage feeding times, and sends information to the information display module 34.
The full load detection module 32 measures the full load condition of the garbage through a sensor 321, and when the measured value reaches a preset threshold value, the full load detection module 32 records and sends a full load signal to the information display module 34, the information voice module 35 and the communication transmission module 36.
The sensor 321 is a laser ranging sensor, an infrared ranging sensor, or an ultrasonic ranging sensor, and performs distance measurement by emitting and feeding back laser light, infrared light, or sound waves. The sensor 321 is preferably a laser ranging sensor.
The sensor 321 detects the full load of the classification trash bin 24 by measuring the height of the volume of the classification trash bin 24 through the sensor 321, and the height of the volume is a threshold value for detecting the full load, wherein the threshold value is 75% to 85% of the height of the classification trash bin 24.
The sensor 321 is arranged along the inner wall of the classification waste bin 23 or perpendicular to the bottom of the classification waste bin 23. The sensor 321 is arranged along the inner wall of the classification garbage can 23, the sensor 321 is arranged at a height threshold value for full-load detection, and the sensor 321 senses the threshold value at 50% ± 10% of the distance between the inner walls of the classification garbage can 23. The sensor 321 is arranged perpendicular to the bottom of the classification garbage can 23, and the sensor 321 sensing threshold is set at the height threshold of the classification garbage can 23.
The information display module 34 includes a display screen 341 and a working status indicator light 342, and the functions of the display screen 341 include displaying the working status of the system, playing videos circularly, displaying the delivery process in real time, displaying the information of specific garbage item names, displaying the information of garbage types, displaying the delivery count of garbage, displaying the garbage amount of the garbage classification box, and displaying the full load information. An operating status indicator light 342 is shown to indicate the operating status of the system unit.
The communication transmission module 36 comprises a serial communication module 361 and a network communication module 362, the serial port communication module 361 realizes the communication among the image acquisition module 11, the image feature processing module 12 and the image recognition module 13, the serial communication module 361 realizes communication between the image recognition module 13 and the delivery mechanism 21, the serial port communication module 361 realizes the communication between the throwing mechanism 21 and the garbage counting module 31, the serial port communication module 361 realizes the communication between the garbage counting module 31 and the information display module 34 and the information voice module 35, the serial communication module 361 enables communication between the sensor 321 and the full load detection module 32, the serial communication module 361 implements the full load detection module 32, the information display module 34, the information voice module 35, and the network communication module 362. The network communication module 362 implements network networking, updates the pre-training model loading unit 132 of the image recognition module 13, and sends the full-load alarm information of the full-load detection module 32.
Based on the system, the embodiment of the garbage classification method is provided, and the intelligent garbage front-end classification method comprises the following steps: after the garbage is placed into the storage device, the garbage image recognition system recognizes the garbage placement and the garbage category, and the automatic garbage classification and delivery system delivers the garbage to the classification garbage can; the garbage information processing system detects the working condition and feeds back information in real time, and the system function is updated in a networking mode.
The method for the system to identify garbage input and garbage categories is as follows.
In step S11, the image acquisition module 11 captures an ith frame image as a reference calibration image through the camera 111. The image acquisition module 11 transmits the reference calibration image to the image feature processing module 12 to extract a key feature value, and further transmits the key feature value to the image recognition module 13 as a reference standard value.
Step S12, the image acquisition module 11 continuously captures images through the camera 111, acquires one frame of image to be identified every n frames of images compared with the i-th frame of reference calibration image, transmits the acquired image to the image feature processing module 12 to extract a key feature value, further transmits the extracted key feature value to the image identification module 13, compares the extracted key feature value with a reference standard value, and determines that garbage is put in if the difference exceeds a set threshold.
In step S13, after the storage device 22 is loaded with garbage, the image recognition module 13 determines that garbage is loaded, and sends a signal to the garbage information processing system 3 and the image recognition module 13. The image recognition module 13 abstracts the bottom-layer features to the high-layer features of the image features to be recognized, and the image feature recognition comparison unit 131 compares the image features with the data loaded by the pre-training model loading unit 132 to obtain the probability value of the image to be recognized belonging to a specific category.
In step S14, the image recognition module 13 takes the highest probability value of the image to be recognized or a value exceeding a certain threshold as the category attribute of the thrown garbage. The image recognition module 13 encodes classification information according to the garbage category attribute, and sends the classification information to the feeding mechanism 21 through the serial communication module 361.
In step S15, after the feeding mechanism 21 completes the execution, the feeding mechanism 21 feeds back a reset signal to the garbage image recognition system 1 through the serial communication module 361. The image acquisition module 11 is cleared, and the image feature processing module 12 is cleared. The image acquisition module 11 captures the ith frame image as a reference calibration image, and the subsequent steps are the same as those of S11-S14.
The method for automatically classifying and releasing the system is specifically as follows.
In step S21, the feeding mechanism 21 receives the code classification information of the image recognition module 13, and drives the bottom layer motor or steering engine 211 and the upper layer motor or steering engine 212 to move jointly.
And step S22, the bottom layer motor or steering engine 211 and the upper layer motor or steering engine 212 move in a combined manner, so that the angle of the object placing device 22 is changed, or/and the phase of the classification garbage can 23 is changed, and classification and throwing of garbage is realized.
Step S23, after the classified garbage is thrown, the bottom layer motor or steering engine 211 and the upper layer motor or steering engine 212 move in a combined manner to restore the initial state of the storage device 22 or/and the classified garbage can 23. The feeding mechanism 21 feeds back a reset signal to the garbage image recognition system 1.
The method for detecting the working condition and feeding back the information in real time by the system is specifically as follows.
Step S31, the information display module 34 and the information voice module 35 of the spam processing system 3 cyclically play the spam classification method and display the recorded data of the spam recording module 33.
Step S32, the garbage image recognition system 1 determines that garbage is thrown in, and the information display module 34 displays the real-time working image of the automatic garbage classification throwing system 2 collected by the garbage image recognition system 1.
In step S33, the counting module 31 is triggered by the operation of the feeding mechanism 21, the counting module 31 records the number of garbage feeding times, the counting module 31 sends information to the garbage information recording module 33 to update the recorded data, and the counting module 31 sends feeding count information to the information display module 34.
In step S34, the information display module 34 reflects the system operating status and information in real time. The display screen 341 displays the working process of the garbage throwing mechanism 21 and the system working state and information stored in the garbage information recording module 33 in real time. The working state indicator light 342 displays the system working state stored in the garbage information recording module 33. The information voice module 35 broadcasts the type to which the garbage belongs.
In step S35, after the classified garbage is thrown, the throwing mechanism 21 feeds back a signal to the garbage information processing system 3, and the garbage information processing system 3 returns to step S31.
In step S36, the release mechanism 21 is operated to trigger the full load detection module 32. The sensor 321 measures sensing points at intervals of a certain time period, and detects whether a measurement threshold is reached.
In step S37, when the detected value exceeds the threshold value, the full-load detection module 32 sends full-load information to the garbage information recording module 33, and the garbage information recording module 33 updates the recorded data.
In step S37, if the detected value exceeds the measurement threshold, the communication transmission module 36 sends the full information of the spam recording module 33 to the server via the network. And cleaning the classification garbage box 23 according to the codes and the corresponding physical addresses of the intelligent garbage front-end classification system, and recovering the initial state of the full-load information of the garbage information recording module 33.
The method for updating the system function by the system networking is concretely as follows.
During the non-working time of the intelligent garbage front-end classification system, the garbage image identification system 1 performs information and data interaction with a remote server end through the communication transmission module 36 through a network. The garbage image recognition system 1 can replace a new network model trained by the server, and update of garbage recognition types and expansion of garbage recognition quantity are realized.
The above examples are only used to illustrate the technical solutions of the present invention, and do not limit the scope of the present invention.
Claims (8)
1. The utility model provides an intelligence rubbish front end classification system which characterized in that includes: the system comprises a garbage image recognition system, an automatic garbage classification and delivery system and a garbage information processing system;
the junk image identification system comprises an image acquisition module, an image feature processing module and an image identification module, wherein the image acquisition module acquires an initial calibration image, a junk image and a system working process image, the image feature processing module extracts an image key feature value, and the image identification module performs high-level abstraction and data comparison on image features and judges an image to be identified;
the automatic garbage classification and throwing system comprises a throwing mechanism, an object placing device and a classification garbage can, wherein the throwing mechanism automatically classifies and throws garbage, the object placing device prestores the garbage, and the classification garbage can classifies and stores the garbage;
the garbage information processing system comprises a garbage counting module, a full-load detection module, a garbage information recording module, an information display module, an information voice module and a communication transmission module, wherein the garbage counting module is used for metering the garbage throwing quantity, the full-load detection module is used for measuring the storage allowance of the classified garbage can, the information recording module is used for storing system parameter information, the information display module is used for displaying the system parameter information, the information voice module is used for broadcasting the system parameter information in a voice mode, and the communication transmission module is used for transmitting information data and pre-training model data;
the garbage information processing system comprises a sensor, a serial communication module and a network communication module, wherein the sensor measures the height of a volume object of a classified garbage can, a certain height of the volume object is used as a threshold value for full load detection, a full load signal is sent out after a measured value reaches a preset threshold value, the serial communication module is used for information data transmission of each module in the system, the network communication module is used for network networking, a pre-training model loading unit of an image recognition module is updated, and full load alarm information of the full load detection module is sent;
the arrangement mode of the sensors is that the sensors are arranged along the inner wall of the classification dustbin or are arranged vertically at the bottom of the classification dustbin, the height threshold is 75% to 85% of the height of the classification dustbin, if the sensors are arranged along the inner wall of the classification dustbin, the sensors are arranged at the height threshold, the sensing distance of the sensors is 50% +/-10% of the distance of the inner wall of the classification dustbin, and if the sensors are arranged vertically at the bottom of the classification dustbin, the sensing distance of the sensors is the height threshold of full load detection;
the image acquisition module captures an ith frame of image as a reference calibration image through the camera, the image acquisition module transmits the reference calibration image to the image feature processing module to extract a key feature value and further transmits the key feature value to the image recognition module as a reference standard value, the image acquisition module continuously captures the image through the camera, compared with the ith frame of reference calibration image, one frame of image to be recognized is acquired at intervals of n frames of images, the image feature processing module extracts the key feature value and further transmits the key feature value to the image recognition module to be compared with the reference standard value, if the difference value exceeds a set threshold value, the garbage is judged to be put in, after the garbage is put in the storage device, the image recognition module judges that the garbage is put in and sends a signal to the garbage information processing system and the image recognition module, and the image recognition module abstracts the features of the image to be recognized from bottom layer features to high-layer features, the image feature recognition comparison unit compares the image feature recognition comparison unit with the data loaded by the pre-training model loading unit to obtain the probability value of the image to be recognized belonging to the specific category, the image recognition module takes the highest probability value or exceeding a certain threshold value of the image to be recognized as the category attribute of the thrown rubbish, and the image recognition module encodes classification information according to the category attribute of the rubbish and sends the classification information to the throwing mechanism through the serial port communication module.
2. The intelligent garbage front-end classification system according to claim 1, further comprising: the device comprises an image characteristic identification comparison unit, a pre-training model loading unit, a camera, a light supplementing device, a light shielding plate or a light shielding cover;
the image characteristic identification comparison unit compares and analyzes the image characteristic value;
the pre-training model loading unit stores and loads a pre-training model and data of the server;
the camera is used for taking real-time image, and light filling device and light screen or shading lid are used for reducing the influence of external strong environment light and eliminate the shadow.
3. The intelligent garbage front-end classification system according to claim 1, further comprising: the bottom motor or the steering engine, the upper motor or the steering engine, the bottom motor or the steering engine and the upper motor or the steering engine are used for changing the angle of the object placing device or/and changing the phase of the classification dustbin.
4. An intelligent garbage front-end classification method, based on any one of claims 1-3, comprising: the method comprises the following steps of identifying a garbage image, automatically classifying and putting garbage, and processing garbage information, wherein the method comprises the following steps:
the system automatically identifies garbage input, collects a frame of image to be identified, compares the image with a reference standard value, judges that garbage is input if the difference value exceeds a set threshold value, further automatically identifies the garbage type, abstracts the characteristics of the image to be identified from the bottom layer characteristics to the high layer characteristics, compares the characteristics with data loaded by a pre-training model loading unit to obtain the class attributes of the garbage, and automatically classifies and puts the garbage into a classification garbage bin;
the system detects the working condition and feeds back information in real time, serial port communication in the system transmits data information, the working process of garbage throwing and system information are displayed in real time, and the system information is broadcasted in a voice mode;
in the non-working time, the system performs information and data interaction with the server through the network, updates and loads the network model and data trained by the server, and updates or expands the type of garbage recognition.
5. The intelligent garbage front-end classification method according to claim 4, further comprising:
the image acquisition module captures an ith frame image as a reference calibration image, and the image feature processing module extracts a key feature value as a reference standard value;
the image acquisition module continuously captures images, one frame of image to be identified is acquired at every n frames of images, the image characteristic processing module extracts key characteristic values, and the image identification module compares the key characteristic values with reference standard values;
if the set threshold value is exceeded, the system recognizes that garbage is put in.
6. The intelligent garbage front-end classification method according to claim 4, characterized in that the intelligent garbage front-end classification method further comprises:
the image recognition module abstracts the bottom-layer features to the high-layer features of the image features to be recognized, and the image feature recognition comparison unit compares the image features with the data of the pre-training model loading unit to obtain the probability value of the image to be recognized belonging to the concrete garbage or the garbage category; and the image recognition module takes the highest probability value or the probability value exceeding a certain threshold value as the category attribute of the thrown rubbish.
7. The intelligent garbage front-end classification method according to claim 4, characterized in that the intelligent garbage front-end classification method further comprises:
the garbage image recognition system codes the garbage category attribute, the garbage category attribute is sent to the throwing mechanism through the serial port communication module, the throwing mechanism drives a motor or a steering engine to move in a combined mode, the angle of the object placing device or/and the phase of the classification garbage can is changed, and garbage is thrown into the classification garbage can.
8. The intelligent garbage front-end classification method according to claim 4, characterized in that the intelligent garbage front-end classification method further comprises:
the garbage throwing mechanism operates to trigger garbage throwing counting, a display screen displays the working process of throwing garbage by the throwing mechanism in real time, the display screen displays the working state and information of the system, and an information voice module broadcasts system information;
the throwing mechanism operates to trigger full load detection, the sensor measures induction points at intervals of a certain time period T, and whether the induction points reach a measurement threshold value is detected;
if the detection value exceeds the measurement threshold value, the full-load detection module sends full-load information to the garbage information recording module through the serial port communication module, and the garbage information recording module updates recorded data;
if the detection value exceeds the measurement threshold value, the full-load detection module sends full-load information to the server side through the network communication module, the classified garbage can is cleaned according to the codes and the corresponding physical addresses of the intelligent garbage front-end classification system, and the full-load information of the garbage information recording module is recovered to the initial state.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110475037.XA CN113135368B (en) | 2021-04-29 | 2021-04-29 | Intelligent garbage front-end classification system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110475037.XA CN113135368B (en) | 2021-04-29 | 2021-04-29 | Intelligent garbage front-end classification system and method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113135368A CN113135368A (en) | 2021-07-20 |
| CN113135368B true CN113135368B (en) | 2022-07-22 |
Family
ID=76816583
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110475037.XA Active CN113135368B (en) | 2021-04-29 | 2021-04-29 | Intelligent garbage front-end classification system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113135368B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114229291A (en) * | 2021-11-25 | 2022-03-25 | 中铁四川生态城投资有限公司 | Garbage transport system, method and computer readable storage medium |
| CN114194648A (en) * | 2021-12-02 | 2022-03-18 | 中智讯（武汉）科技有限公司 | Garbage classification method and garbage classification system based on artificial intelligence |
| CN114821440B (en) * | 2022-05-12 | 2023-03-24 | 清研灵智信息咨询(北京)有限公司 | Mobile video stream content identification and analysis method based on deep learning |
| CN114782900A (en) * | 2022-06-21 | 2022-07-22 | 成都泽赛普石油工程技术服务有限公司 | Target monitoring and processing system and method based on Yolov5 algorithm |
| CN115610868A (en) * | 2022-11-09 | 2023-01-17 | 江苏华旭环卫科技有限公司 | Wisdom classification system based on thing networking |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101477254B1 (en) * | 2014-06-13 | 2014-12-30 | 엠엔테크(주) | Garbage collection device for automatic pick-up equipment |
| CN109800772A (en) * | 2019-01-30 | 2019-05-24 | 广州市载道信息科技有限公司 | A kind of data identification method |
| CN212314520U (en) * | 2020-09-07 | 2021-01-08 | 广州东软科技有限公司 | Garbage mixed feeding or subpackaging device |
| CN112298844A (en) * | 2019-07-29 | 2021-02-02 | 杭州海康威视数字技术股份有限公司 | Garbage classification monitoring method and device |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107499763A (en) * | 2017-08-18 | 2017-12-22 | 四川壹家易信息技术有限公司 | Side is provided with the garbage transporting equipment of transmission belt |
| CN109389161B (en) * | 2018-09-28 | 2021-08-31 | 广州大学 | Garbage identification evolutionary learning method, device, system and medium based on deep learning |
| CN110294243A (en) * | 2019-08-02 | 2019-10-01 | 重庆住派科技有限公司 | Store robot |
| CN110723432A (en) * | 2019-09-20 | 2020-01-24 | 精锐视觉智能科技（深圳）有限公司 | Garbage classification method and augmented reality equipment |
| CN110893956A (en) * | 2019-11-18 | 2020-03-20 | 山东科技大学 | Intelligent garbage classification and management system, garbage classification identification method and self-inspection method |
| CN111444813A (en) * | 2020-03-23 | 2020-07-24 | 北京字节跳动网络技术有限公司 | Method, device, equipment and storage medium for identifying attribute classification of target object |
-
2021
- 2021-04-29 CN CN202110475037.XA patent/CN113135368B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101477254B1 (en) * | 2014-06-13 | 2014-12-30 | 엠엔테크(주) | Garbage collection device for automatic pick-up equipment |
| CN109800772A (en) * | 2019-01-30 | 2019-05-24 | 广州市载道信息科技有限公司 | A kind of data identification method |
| CN112298844A (en) * | 2019-07-29 | 2021-02-02 | 杭州海康威视数字技术股份有限公司 | Garbage classification monitoring method and device |
| CN212314520U (en) * | 2020-09-07 | 2021-01-08 | 广州东软科技有限公司 | Garbage mixed feeding or subpackaging device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113135368A (en) | 2021-07-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113135368B (en) | Intelligent garbage front-end classification system and method | |
| CN110053903B (en) | Household garbage classification and recovery management system and method | |
| US20230196307A1 (en) | System and method for waste management | |
| CN110436083B (en) | Intelligent household combined garbage can, system and method | |
| US20200082354A1 (en) | Smart waste management system | |
| CN110697273A (en) | Intelligent household garbage identification and automatic classification system and method based on iterative learning control | |
| CN104495139B (en) | Garbage dropping box and class-based dropping method of garbage | |
| CN107282477A (en) | Plastic garbage classification retracting device and control method based on image recognition technology | |
| CN111583508A (en) | Garbage classification management system | |
| CN110861851A (en) | Community garbage classification system and method based on Internet of things | |
| CN206546593U (en) | House refuse intelligent classification reclaims cloud identifying system | |
| CN111846693A (en) | Intelligent garbage classification putting supervision and control system based on real-name system | |
| CN210339124U (en) | Intelligent classification dustbin with automatically, allocate function | |
| CN204916794U (en) | Intelligence waste classification device and system | |
| CN107123076A (en) | A kind of waste management system and waste management method | |
| CN112407655A (en) | Garbage classification and recovery system based on Internet of things | |
| CN110990633A (en) | Garbage classification fixed-point management system and method based on image retrieval and public system | |
| CN111056191A (en) | Intelligent garbage collection system capable of classifying and delivering multiple garbage | |
| CN112607253A (en) | Kitchen garbage classification recycling credit evaluation system based on big data | |
| CN111806914A (en) | Intelligent garbage classification and recovery system for community | |
| CN112241651A (en) | Data display method and system, data processing method, storage medium and system | |
| CN211894657U (en) | Intelligent garbage classification station | |
| CN113978966B (en) | System and method for accurately collecting kitchen garbage and tracking sources | |
| CN110775483A (en) | Community intelligent garbage recycling device and method, electronic equipment and storage medium | |
| CN111661509B (en) | Garbage recycling and classifying management system based on interaction of Internet and Internet of things |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |