CN111797758A - Identification and positioning technology for plastic bottles - Google Patents
Identification and positioning technology for plastic bottles Download PDFInfo
- Publication number
- CN111797758A CN111797758A CN202010630321.5A CN202010630321A CN111797758A CN 111797758 A CN111797758 A CN 111797758A CN 202010630321 A CN202010630321 A CN 202010630321A CN 111797758 A CN111797758 A CN 111797758A
- Authority
- CN
- China
- Prior art keywords
- module
- algorithm
- chip
- plastic bottles
- identification
- 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
- G06V20/20—Scenes; Scene-specific elements in augmented reality scenes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T1/00—General purpose image data processing
- G06T1/20—Processor architectures; Processor configuration, e.g. pipelining
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Multimedia (AREA)
- Processing Of Solid Wastes (AREA)
Abstract
The invention discloses a recognition and positioning technology for plastic bottles, which comprises two parts, namely a hardware part and a software system. The hardware part comprises an MCU, a power management module, an image acquisition device, a touch screen and a peripheral extension module. The software part comprises the steps of taking a picture, preprocessing the picture, training a model, obtaining a required model file, testing whether the identification precision of the model file meets the expected requirement, adjusting parameters related to an algorithm according to an experimental result, and verifying and comparing. Thereby realizing positioning identification. The invention mainly aims to realize intelligent garbage classification and save labor cost, thereby maintaining sustainable development of ecological environment.
Description
Technical Field
The invention belongs to the aspect of environmental protection, and relates to an intelligent garbage sorting-plastic bottle identification and positioning technology.
Background
Along with the increasing living standard of people, how to treat a large amount of domestic garbage becomes a great problem facing people at present. The conventional garbage classification recycling is to classify the garbage firstly when the garbage is put in initially, put the garbage of the same type into a designated garbage can, convey the garbage to a garbage transfer station, and sort out the recyclable garbage according to secondary classification of the special garbage by workers for recycling. This is both time consuming and laborious. Therefore, the invention of the intelligent garbage sorting system is particularly important.
At present, domestic garbage treatment modes mostly adopt a traditional method, namely a mode of a garbage recycling station. The green garbage can is cleaned and transported by a garbage collecting and transporting vehicle, and the black garbage can is collected and transported to a transfer station of each area by a cleaner of the property or a cleaner of an environmental protection department. Foreign waste sorting is in the research and development stage and has a period of time away from industrial production application because waste sorting is a rather complicated environment, and has many articles and is in a state of being polluted and deformed. In the intelligent classification technology, the machine vision and positioning technology are very important, a key technology can be provided for the research of a large number of garbage classifications, and the intelligent classification technology has long-term significance for environmental protection and economic development.
Disclosure of Invention
The invention provides an intelligent garbage sorting system based on a YOLO algorithm, which has great advantages of garbage classification based on the YOLO algorithm of deep learning and aims to solve the problems of low speed, difficult training and low accuracy in the existing algorithm. The specific scheme is as follows:
in a first aspect, an example of the present application provides an intelligent garbage sorting method, including:
data sets were captured by a CMOS industrial camera, U300, and acquired by web crawlers, using labelImg for data tagging.
And constructing Darknet, Tensorflow framework and CUDA parallel computing framework for receiving and processing data. And carrying out image recognition according to algorithms such as training models, matching extraction and the like. And positioning the identified plastic bottles according to a detection algorithm and a positioning algorithm.
The system carries out preprocessing on the input picture, the preprocessing operation is mainly carried out in a data enhancement mode, the normalization operation and other operations are carried out on the data set, and convenience is provided for subsequent training.
And training the model, namely installing the environment required by the system before training the model, and starting training after the test is correct. The main process is to feed the preprocessed data set into the neural network of YOLOv 3.
In a second aspect, the present application provides an intelligent garbage sorting system, including:
hardware system: the system consists of a camera for collecting images of the plastic bottle and an ARM processor for operating an identification algorithm.
And (3) algorithm research: mainly relates to an algorithm for identifying an image of a plastic bottle and positioning the position of the plastic bottle.
A software system: the method mainly comprises code implementation of a research algorithm and codes of code transplantation and an interactive interface on an ARM board, and a software system runs under Linux.
The operation flow of the whole system is as follows: 1. and setting the position for placing the camera to correct the camera. 2. And carrying out image recognition according to algorithms such as training models, matching extraction and the like. 3. And positioning the identified plastic bottles according to a detection algorithm and a positioning algorithm.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below, it is obvious that the drawings in the following description are only embodiments of the present application, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
Fig. 1 is a schematic diagram of an overall framework of an intelligent garbage recognition method system provided by an embodiment of the present application.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
The system hardware structure diagram is shown in FIG. 2, which includes
MCU: an ARM processor is selected, and an Exynos4412 chip produced by Samsung is selected in the ARM processor, and a core used is Cortex-A9.
A power management module: because the whole system is large and the required power consumption is large, a PN8370 power supply chip is selected in the design, the chip has output short-circuit protection and overcurrent protection pins, and the average power of the chip can reach more than 78.70%.
An image acquisition device: the system adopts a U300 CMOS industrial camera, a lens is added at the front end of the camera to eliminate perspective errors, and the adjustable range of a focal head is 6-12 mm. The internal integration level is high, the power supply requirement voltage range is smaller, the standard USB2.0 communication protocol is used in the communication protocol, the resolution ratio can be controlled through a program, and the design requirement is adapted.
A touch screen: the design selects a 7-inch LCD screen capacitive screen, and the adopted data signal transmission communication mode is LVDS.
A peripheral expansion module: in order to debug the system conveniently, a plurality of peripheral modules are added on the PCB bottom plate, so that the normal operation probability of the system is ensured, and the problems in intermediate links are reduced. The modules comprise a necessary reset circuit module, a BOOT mode selection module, various communication transmission modules such as a JATG module, a network card module, a USB module and the like, and a necessary TF card/SD card module for reading and writing a storage module.
The overall design scheme of the software system is shown in fig. 3, which mainly comprises the following steps:
step 1: the position of the camera is set and adjusted.
Step 2: and preprocessing the shot picture, namely normalizing the data set in a data enhancement mode.
And step 3: and (3) training a model, installing a CUDNN acceleration library, and obtaining a required model file through continuous iterative training by using a Yolov3 algorithm.
And 4, step 4: whether the identification precision of the model file meets the expected requirement is tested, and verification and comparison are carried out according to parameters related to an experimental result adjusting algorithm. And transplanting the intelligent garbage sorting system to an ARM to realize plastic bottle identification and positioning which are key technical researches of the intelligent garbage sorting system.
The system software flow chart is shown in fig. 4.
Claims (2)
1. An identification and location technique for plastic bottles, comprising: the system comprises an MCU, a power management module, image acquisition equipment, a touch screen and a peripheral extension module;
the MCU is selected from an ARM processor, an Exynos4412 chip produced by Samsung is selected from the ARM processor, and a core used is Cortex-A9;
the power management module selects a PN8370 power chip, the chip has output short-circuit protection and overcurrent protection pins, and the average power of the chip can reach more than 78.70%;
the image acquisition equipment adopts a U300 CMOS industrial camera, a lens is added at the front end of the camera to eliminate perspective errors, the adjustable range of a focal head is 6-12 mm, and the internal integration level is high;
the touch screen adopts a 7-inch LCD screen capacitive screen, the adopted data signal transmission communication mode is LVDS, the peripheral extension module comprises a necessary reset circuit module, a BOOT mode selection module, various communication transmission modules such as a JATG module, a network card module, a USB module and the like, and a necessary TF card/SD card module for reading and writing storage modules, and the modules mainly aim to ensure the normal operation probability of the system and reduce the occurrence of problems in intermediate links.
2. The system of claim 1, wherein the plastic bottles are identified and located accurately and rapidly, the algorithm selected for this purpose is YOLOV3 algorithm, the network model in the algorithm uses convolutional neural network and is evaluated on Pascal VOC detection data set, the prediction class probability and bounding box coordinates of the last layer in YOLO algorithm are normalized by the input image width and height, the bounding box width and height are also normalized to make its value fall within [0,1] interval, the linear activation function is used for the last layer, and all other layers are linearly activated using the following modification as formula 1:
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010630321.5A CN111797758A (en) | 2020-07-03 | 2020-07-03 | Identification and positioning technology for plastic bottles |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010630321.5A CN111797758A (en) | 2020-07-03 | 2020-07-03 | Identification and positioning technology for plastic bottles |
Publications (1)
Publication Number | Publication Date |
---|---|
CN111797758A true CN111797758A (en) | 2020-10-20 |
Family
ID=72809910
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010630321.5A Pending CN111797758A (en) | 2020-07-03 | 2020-07-03 | Identification and positioning technology for plastic bottles |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111797758A (en) |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN205232360U (en) * | 2015-11-26 | 2016-05-11 | 浙江工业大学 | Wireless transmission high-definition video monitoring system |
CN110219271A (en) * | 2019-06-26 | 2019-09-10 | 苏州奥创智能科技有限公司 | A kind of garbage classification intelligent vision system and method applied to Automatic vehicle for cleaning road |
CN110321853A (en) * | 2019-07-05 | 2019-10-11 | 杭州巨骐信息科技股份有限公司 | Distribution cable external force damage prevention system based on video intelligent detection |
US20200005468A1 (en) * | 2019-09-09 | 2020-01-02 | Intel Corporation | Method and system of event-driven object segmentation for image processing |
CN110781896A (en) * | 2019-10-17 | 2020-02-11 | 暨南大学 | Track garbage identification method, cleaning method, system and resource allocation method |
CN110796186A (en) * | 2019-10-22 | 2020-02-14 | 华中科技大学无锡研究院 | Dry and wet garbage identification and classification method based on improved YOLOv3 network |
US20200175311A1 (en) * | 2018-11-29 | 2020-06-04 | Element Ai Inc. | System and method for detecting and tracking objects |
-
2020
- 2020-07-03 CN CN202010630321.5A patent/CN111797758A/en active Pending
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN205232360U (en) * | 2015-11-26 | 2016-05-11 | 浙江工业大学 | Wireless transmission high-definition video monitoring system |
US20200175311A1 (en) * | 2018-11-29 | 2020-06-04 | Element Ai Inc. | System and method for detecting and tracking objects |
CN110219271A (en) * | 2019-06-26 | 2019-09-10 | 苏州奥创智能科技有限公司 | A kind of garbage classification intelligent vision system and method applied to Automatic vehicle for cleaning road |
CN110321853A (en) * | 2019-07-05 | 2019-10-11 | 杭州巨骐信息科技股份有限公司 | Distribution cable external force damage prevention system based on video intelligent detection |
US20200005468A1 (en) * | 2019-09-09 | 2020-01-02 | Intel Corporation | Method and system of event-driven object segmentation for image processing |
CN110781896A (en) * | 2019-10-17 | 2020-02-11 | 暨南大学 | Track garbage identification method, cleaning method, system and resource allocation method |
CN110796186A (en) * | 2019-10-22 | 2020-02-14 | 华中科技大学无锡研究院 | Dry and wet garbage identification and classification method based on improved YOLOv3 network |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN112395978B (en) | Behavior detection method, behavior detection device and computer readable storage medium | |
CN110598752B (en) | Image classification model training method and system for automatically generating training data set | |
CN110472082B (en) | Data processing method, data processing device, storage medium and electronic equipment | |
CN105389593A (en) | Image object recognition method based on SURF | |
CN104021375A (en) | Model identification method based on machine learning | |
CN105260734A (en) | Commercial oil surface laser code recognition method with self modeling function | |
CN103065122A (en) | Facial expression recognition method based on facial motion unit combination features | |
CN105427320A (en) | Image segmentation and extraction method | |
CN102034096A (en) | Video event recognition method based on top-down motion attention mechanism | |
CN111767908B (en) | Character detection method, device, detection equipment and storage medium | |
CN111797239A (en) | Application program classification method and device and terminal equipment | |
CN109816634B (en) | Detection method, model training method, device and equipment | |
CN112733936A (en) | Recyclable garbage classification method based on image recognition | |
CN105956096A (en) | Rapid question searching method and device based on mobile terminal | |
CN114138784A (en) | Information tracing method and device based on storage library, electronic equipment and medium | |
CN110766045A (en) | Underground drainage pipeline disease identification method, intelligent terminal and storage medium | |
CN111797758A (en) | Identification and positioning technology for plastic bottles | |
Zhang et al. | An automatic defect detection method for TO56 semiconductor laser using deep convolutional neural network | |
Peng et al. | Application of deep residual neural network to water meter reading recognition | |
CN108345943B (en) | Machine learning identification method based on embedded coding and contrast learning | |
CN112527179B (en) | Scribble image recognition method and device and terminal equipment | |
CN105718972B (en) | A kind of information intelligent acquisition method | |
CN111783891A (en) | Customized object detection method | |
Prasad et al. | Trends in handwriting recognition | |
CN113378627A (en) | Automatic garbage classification method and device based on DSP chip and single chip microcomputer control |
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 | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20201020 |
|
WD01 | Invention patent application deemed withdrawn after publication |