CN107514978B - Recovery machine packing material shape identification method and system based on infrared grating detection - Google Patents
Recovery machine packing material shape identification method and system based on infrared grating detection Download PDFInfo
- Publication number
- CN107514978B CN107514978B CN201710517274.1A CN201710517274A CN107514978B CN 107514978 B CN107514978 B CN 107514978B CN 201710517274 A CN201710517274 A CN 201710517274A CN 107514978 B CN107514978 B CN 107514978B
- Authority
- CN
- China
- Prior art keywords
- data
- bottle
- module
- infrared grating
- packing
- 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
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
- G01B11/24—Measuring arrangements characterised by the use of optical techniques for measuring contours or curvatures
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Processing Of Solid Wastes (AREA)
Abstract
The invention provides a method and a system for identifying the shape of a packaging material of a recycling machine based on infrared grating detection, wherein a network service platform comprises the following components: the system comprises a platform network communication module, a packing shape feature configuration file generation module, a recycling machine configuration module and a platform alarm module; the recovery machine terminal module includes: the system comprises an infrared grating state detection module, a packing object identification module, a packing object throwing rebate module, a terminal alarm detection module, a terminal network communication module and a terminal maintenance management module; the infrared grating acquisition system includes: the system comprises a direct current motor, a belt transmission device, an infrared grating acquisition module and an infrared grating communication module. Has the advantages that: (1) the infrared grating device is adopted to detect the shape and the size of the packaged object, so that the sensitivity and the precision are high; in addition, the infrared grating device is not easily influenced by illumination factors, and has high stability, long service life and low maintenance cost. (2) The feature extraction algorithm of the invention has stronger robustness and fault-tolerant capability to noise.
Description
Technical Field
The invention belongs to the technical field of intelligent recognition terminals and resource recovery, and particularly relates to a recoverer packing material shape recognition method and system based on infrared grating detection.
Background
Since the climate change conference of the Copenhagen United nations, the handling of climate change becomes a focus problem of international politics, and mainly focuses on the issues of emission reduction target, carbon tax, capital, technical cooperation and the like. China, as a developing large country with high economic growth, needs to cope with higher emission reduction targets proposed by developed countries on the one hand, and needs to maintain rapid economic growth on the other hand. In order to complete the carbon emission reduction target and strive for the carbon emission right, the government of China starts various works from the aspects of legislation, mobilization, quota allocation and the like, and a mature carbon trading market system is gradually established. In the future, the relation between the reduction of carbon emission and enterprises and individuals is tighter, the enterprises with surplus capacity and high carbon emission bear greater carbon emission responsibility, the low-carbon and environment-friendly enterprises are favored by the market, and the development of low-carbon and environment-friendly economy is promoted continuously.
Compared with the original resource, the renewable resource recycling has the advantages of reducing resource consumption, reducing production cost, reducing environmental pollution and carbon emission, and is the key for developing circular economy. The standard packing materials, such as various packing materials, have the characteristics of continuous generation in a large amount of use, low recycling value, certain pollution and the like. The standard packing materials are recycled, so that the environmental pollution and the carbon emission can be reduced, and the waste can be changed into valuable.
The standard package recovery route is mainly as follows: the individual waste materials are collected to a waste material recovery company through fixed or mobile individual waste material recovery points in the community, and then are dispersed to enterprises who take the waste materials as raw materials for production. Often, a large number of illegal operating points which are not officially registered exist in the recovery systems and organizations, when standard packing materials are processed, secondary pollution can be caused to the environment, and articles produced by using the standard packing materials have greater hidden dangers to the society.
At present, most of intelligent recycling machines on the market use bar code identification and image identification as standard packing object identification modes, and then classify the standard packing objects and recycle according to preset recycling prices. The invention discloses an intelligent packaging material recycling machine, and provides a roller type bar code recognition recycling device in the patent application No. 201510876046.4. The device can't effectively retrieve the packing material that bar code department flattened and the bar code is damaged to can't solve cover bar code and throw cheating such as incomplete bottle and throw the bottle problem of throwing.
The invention discloses a Chinese patent application, namely a recovery device based on an image recognition device, and the patent of the application 201220226902.3 provides a packaged object recognition device for image acquisition and processing. The system has the following disadvantages: the image acquisition has higher requirements on a camera, a light source and working conditions, and the production and maintenance cost is high; the image processing part needs to be configured with a high-performance processor to meet the real-time requirement; illumination, background color and noise easily influence imaging, so that the robustness of the system is poor; the packing materials usually have differences such as deformation and lost evaluation marks, and the corresponding images have different characteristics such as outlines, colors and the like, so that the high recognition rate is difficult to maintain.
Therefore, how to effectively solve the problems is an urgent matter at present.
Disclosure of Invention
Aiming at the defects in the prior art, the invention provides a recycling machine packing material shape identification method and system based on infrared grating detection, which can effectively solve the problems.
The technical scheme adopted by the invention is as follows:
the invention provides a recycling machine packing material shape recognition method based on infrared grating detection, which comprises the following steps of:
step 1, generating a configuration file of the shape and the characteristics of a packaged object, and the steps are as follows:
step 1.1, configuring a bar code gun and an infrared grating acquisition device on a terminal recycling machine; the terminal recycling machine scans the bar code data of the packing materials by adopting the bar code gun; meanwhile, the terminal recycling machine adopts the infrared grating collecting device to collect N pieces of infrared grating data of the packing materials; wherein N is a natural number; the N pieces of infrared grating data are size data of N different parts of the packing material;
step 1.2, the terminal recycling machine uploads the bar code data of the packaged object and the corresponding N pieces of infrared grating data to a network service platform;
step 1.3, the network service platform performs feature extraction on the N pieces of infrared grating data corresponding to the packaging object bar code data, and maps each piece of infrared grating data to a feature vector of a packaging object feature vector space, so as to obtain a packaging object feature vector space consisting of N feature vectors;
step 1.4, rejecting abnormal feature vectors deviating from the average value to be too large according to a statistical principle for the space of the feature vectors of the packing objects, calculating the average value and the standard deviation of the feature values of the normal feature vectors according to attributes, and obtaining the average value and the normal variation range of the features of the packing objects so as to obtain the shape feature data of the packing objects corresponding to the bar code data of the packing objects; wherein the shape characteristic data of the packing object is the mean value and the normal variation range of the characteristics of the packing object;
step 1.5, continuously extracting the characteristics of the next packing material by adopting the mode of the step 1.1 to the step 1.4, and continuously circulating in such a way to obtain a plurality of packing material bar code data, wherein each packing material bar code data corresponds to the shape characteristic data of the packing material;
storing the bar code data of each packing object and the corresponding shape characteristic data of the packing object into a configuration file, thereby obtaining the configuration file of the shape characteristic of the packing object;
step 2, the actual recognition process of the shape of the packing object comprises the following steps:
step 2.1, the terminal recycling machine receives the configuration file of the shape and the characteristics of the packaged objects issued by the network service platform and loads the configuration file of the shape and the characteristics of the packaged objects into a memory;
step 2.2, when the terminal recycling machine is put into a package, the terminal recycling machine scans the package bar code data of the package by adopting the bar code gun; meanwhile, the terminal recycling machine adopts the infrared grating collecting device to collect N pieces of infrared grating data of the packing materials;
step 2.3, the terminal recycling machine performs feature extraction on the N pieces of infrared grating data acquired in the step 2.2 to obtain a packaging object feature vector space consisting of N feature vectors;
step 2.4, the terminal recycling machine judges whether the bar code gun successfully scans and obtains the bar code data of the packing object in the step 2.2; if the code scanning is successful, executing the step 2.5-the step 2.6; if the code scanning fails, executing step 2.7;
step 2.5, the terminal recycling machine further judges whether the obtained packaged object bar code data exists in the packaged object shape feature configuration file, if not, the packaged object bar code data obtained in the step 2.4 and the packaged object feature vector space obtained in the step 2.3 are uploaded to the network service platform, and the network service platform updates the packaged object shape feature configuration file; if so, performing step 2.6;
step 2.6, the terminal recycling machine searches the configuration file of the shape characteristics of the packing object to obtain the shape characteristic data of the packing object corresponding to the bar code data of the packing object obtained in the step 2.4; the shape characteristic data of the packing object is the mean value and the normal variation range of the characteristic of the packing object;
then, the terminal recycling machine judges whether the characteristic vector space of the packing object obtained in the step 2.3 is matched with the searched shape characteristic data of the packing object, and if the matching is successful, the packing object is recycled; if the matching is unsuccessful, rejecting to recycle the wrappage;
step 2.7, the terminal recovery machine extracts a plurality of items of packing object shape characteristic data with the highest historical packing object throwing frequency from the packing object shape characteristic configuration file;
then, the characteristic vector space of the packing materials obtained in the step 2.3 is sequentially matched with the shape characteristic data of each packing material, and if the matching is successful, the packing materials are recycled; if the matching is not successful, the package is rejected for recycling.
Preferably, in step 1.1, the size data of the N different locations includes: bottle cap size data, bottle head size data, bottle tail size data, bottle body size data and volume characteristic data.
The invention also provides a recycling machine packaging material shape recognition system based on infrared grating detection, which comprises a network service platform, a recycling machine terminal module and an infrared grating acquisition system;
the network service platform comprises: the system comprises a platform network communication module, a packing shape feature configuration file generation module, a recycling machine configuration module and a platform alarm module;
the platform network communication module is used for communication between the network service platform and the recycling machine terminal module, and receiving and storing the packaging object bar code data uploaded by the recycling machine terminal module and the corresponding N pieces of infrared grating data;
the packaging shape feature configuration file is used for carrying out feature extraction on the N pieces of infrared grating data which are received by the platform network communication module and correspond to the packaging bar code data to obtain a packaging shape feature configuration file;
the recycling machine configuration module is used for configuring relevant configuration information of a recycling machine terminal; the relevant configuration information comprises a recycling machine terminal number, longitude and latitude, an address and a configuration file;
the platform alarm module is used for processing and analyzing alarm information of the recovery machine terminal;
the recycling machine terminal module includes: the system comprises an infrared grating state detection module, a packing object identification module, a packing object throwing rebate module, a terminal alarm detection module, a terminal network communication module and a terminal maintenance management module;
the infrared grating state detection module is used for detecting whether the state of the infrared grating is normal or not;
the package identification module is used for distinguishing packages thrown into the recycling machine;
the package throwing rebate module is used for recording the number of packages thrown by a user and specific package parameter information, and uploading the package parameter information to the network service platform to realize package throwing rebate of the user;
the terminal alarm detection module is used for detecting the current running state of the recovery machine and reporting to the network service platform in an alarm mode when abnormality or hidden danger is found;
the terminal network communication module is used for communicating with the network service platform and uploading the infrared grating data acquired by the infrared grating to the network service platform; receiving a configuration file issued by the network service platform;
the terminal maintenance management module is used for realizing the checking of a maintainer on the current state of the recovery machine and the control function of the recovery machine;
the infrared grating acquisition system comprises: the system comprises a direct current motor, a belt transmission device, an infrared grating acquisition module and an infrared grating communication module;
the direct current motor and the belt transmission device are used for controlling the length and the stability of data acquired by the infrared grating acquisition module;
the infrared grating acquisition module can realize self-checking and resetting of a power-on environment and acquire data of the size of a grating shielding object in an interruption mode;
and the infrared grating communication module is used for realizing data communication between the infrared grating acquisition module and the recovery machine terminal module.
The method and the system for identifying the shape of the packaging material of the recovery machine based on the infrared grating detection have the following advantages:
(1) the infrared grating device is adopted to detect the shape and size of the packaged object, so that the sensitivity and the precision are high, and the accuracy of feature extraction can be ensured; in addition, the infrared grating device is not easily influenced by illumination factors, and has high stability, long service life and low maintenance cost.
(2) The feature extraction algorithm has stronger robustness and fault-tolerant capability on noise, has small calculated amount, and can improve the accuracy of identifying the shape and the size of a packaged object and ensure the real-time property; the configuration file of the shape characteristics of the packaged objects is obtained according to the statistical principle, abnormal data which are not in accordance with the characteristics of the bar codes are eliminated, and errors generated by a motor and a conveyor belt device are eliminated.
Drawings
FIG. 1 is a schematic structural diagram of a recoverer packing material shape recognition system based on infrared grating detection provided by the invention;
FIG. 2 is a schematic diagram of a process for generating a configuration file of shape characteristics of a package according to the present invention;
FIG. 3 is a schematic flow chart of a recoverer packaged object shape identification method based on infrared grating detection provided by the invention;
fig. 4 is a schematic diagram of the calculation of the volume characteristics of the packages provided by the present invention.
Detailed Description
In order to make the technical problems, technical solutions and advantageous effects solved by the present invention more clearly apparent, the present invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
The invention provides a recoverer packaging material shape recognition method and system based on infrared grating detection, aiming at solving the problems that the existing packaging material recoverer cannot realize the purposes of high recovery rate, real-time response and high stability and can fundamentally solve the problem of cheating bottle throwing, and aiming at realizing the following purposes: (1) the production and maintenance costs are reduced. The small grating with the grating W (width) H (height) of 25cm 40cm is adopted, the batch customization cost is low, the service life of the grating is long, the installation is simple, and the maintenance cost is low. (2) The real-time performance is improved. The data volume of the grating collection package is within 100 bytes, and the Android multithreading is utilized to collect and match the infrared grating data, so that the ms-level response speed is achieved. (3) And the anti-interference capability is enhanced. The data acquisition is less restricted by environmental factors, the interference generated by the environment is eliminated, and the robustness of the system is improved. (4) High recognition rate. Under the conditions of deformation, label evaluation loss and the like of the packing materials, the packing materials can still be thrown and distributed for recycling according to the deformation acceptance and the historical common product types. (5) The problem of cheating package throwing is solved. The shape of the packing material is adopted for recognition, so that the problem of fraudulent bottle throwing such as label evaluation, bottle throwing only label and half bottle throwing is effectively solved.
The invention provides a recycling machine packing material shape recognition method based on infrared grating detection, wherein the packing material can be various beverage bottles and other articles, and the method comprises the following steps with reference to figure 3:
step 1, generating a configuration file of the shape and the characteristics of the packaged object, referring to fig. 2, and comprising the following steps:
step 1.1, configuring a bar code gun and an infrared grating acquisition device on a terminal recycling machine; the terminal recycling machine scans the bar code data of the packing materials by adopting the bar code gun; meanwhile, the terminal recycling machine adopts the infrared grating collecting device to collect N pieces of infrared grating data of the packing materials; wherein N is a natural number; the N pieces of infrared grating data are size data of N different parts of the packing material;
step 1.2, the terminal recycling machine uploads the bar code data of the packaged object and the corresponding N pieces of infrared grating data to a network service platform;
step 1.3, the network service platform performs feature extraction on the N pieces of infrared grating data corresponding to the packaging object bar code data, and maps each piece of infrared grating data to a feature vector of a packaging object feature vector space, so as to obtain a packaging object feature vector space consisting of N feature vectors;
step 1.4, rejecting abnormal feature vectors deviating from the average value to be too large according to a statistical principle for the space of the feature vectors of the packing objects, calculating the average value and the standard deviation of the feature values of the normal feature vectors according to attributes, and obtaining the average value and the normal variation range of the features of the packing objects so as to obtain the shape feature data of the packing objects corresponding to the bar code data of the packing objects; wherein the shape characteristic data of the packing object is the mean value and the normal variation range of the characteristics of the packing object;
step 1.5, continuously extracting the characteristics of the next packing material by adopting the mode of the step 1.1 to the step 1.4, and continuously circulating in such a way to obtain a plurality of packing material bar code data, wherein each packing material bar code data corresponds to the shape characteristic data of the packing material;
storing the bar code data of each packing object and the corresponding shape characteristic data of the packing object into a configuration file, thereby obtaining the configuration file of the shape characteristic of the packing object;
more specifically, step 1.1 to step 1.4 include:
1) extracting N pieces of infrared grating data corresponding to the bar code, and converting the N pieces of infrared grating data into N feature vectors in a vector space consisting of a bottle cap, a bottle head, a bottle tail, a bottle body and volume features. The algorithm for extracting the shape characteristics of the packaged objects comprises the following specific steps:
1.1) eliminating distortion data points. Storing the infrared grating data in the data, subtracting adjacent values in the data array to obtain an absolute value, if the absolute value is greater than 10(1 cm), then a distortion point exists at the position, if the left side and the right side of the point are continuously changed and the average values of the left side and the right side are approximately equal, replacing the value of the point with the average values of the left side and the right side, otherwise, selecting the point with the minimum difference between the average values of the left side and the right side and the point to replace the value of the point by using a nearest neighbor rule.
1.2) bottle cap features. The bottle cap is structurally characterized in that the difference between the outer ring of the bottle cap and the bottle neck is detected, and the difference change can be detected by adopting a high-precision grating. And (4) storing the bottle cap data in cap _ data [ ], wherein a sudden decrease change occurs from the ith point, and an increasing change rule occurs after the point. And under the condition that the grating acquisition frequency and the belt transmission speed are fixed, averaging the first m in the cap _ data to obtain the width of the bottle cap.
1.3) bottle head characteristics. The bottle head is characterized in that after the bottle cap is covered, an increasing change rule appears. The raster data generally presents a continuously increasing rule, and the change rate of the raster data is represented by the ratio of the mean value of the same point number of the two sections.
1.4) bottle end characteristics. The bottle tail is characterized in that the width of the bottle tail changes with different bottles, and the average value of a certain length of the bottle tail is calculated to be used as the width of the bottle tail.
1.5) body characteristics. The width of the bottle body reflects the whole width of the bottle, and the average value of the width from the bottle head to the bottle tail is calculated and used as the width of the bottle body.
1.6) volume characteristics. The volume characteristics are used as a rebate basis, and the accurate calculation of the volume of the bottle is very critical. As shown in fig. 4, the hatched portion indicates a cut surface in the belt direction. The volume calculation formula is:
Diwidth of the package in section i, Δ ziThe length between the sections, the grating acquisition frequency and the conveyor speed are timed, Δ ziApproximately constant, and n is the length of data acquired by the grating.
2) According to the statistical principle, calculating the mean value and the standard deviation of different attribute characteristics of the N characteristic vectors according to the attribute types, and removing the characteristic vectors which deviate from the mean value to be too large from the N characteristic vectors by using a 3 sigma rule to obtain M vectors which are corresponding to the bar codes and can truly describe the bottle characteristics.
3) And solving and counting the mean value and the standard deviation of each characteristic attribute of the M characteristic vectors according to the attribute types to obtain a stable point and an acceptance domain of the bottle in a vector space, and taking the stable point and the acceptance domain of each attribute characteristic as the characteristic vectors in the bottle shape library so as to judge whether the characteristic vector of one bottle conforms to the corresponding vector characteristic in the bottle shape library.
Step 2, the actual recognition process of the shape of the packing object comprises the following steps:
step 2.1, the terminal recycling machine receives the configuration file of the shape and the characteristics of the packaged objects issued by the network service platform and loads the configuration file of the shape and the characteristics of the packaged objects into a memory;
step 2.2, when the terminal recycling machine is put into a package, the terminal recycling machine scans the package bar code data of the package by adopting the bar code gun; meanwhile, the terminal recycling machine adopts the infrared grating collecting device to collect N pieces of infrared grating data of the packing materials;
step 2.3, the terminal recycling machine performs feature extraction on the N pieces of infrared grating data acquired in the step 2.2 to obtain a packaging object feature vector space consisting of N feature vectors;
step 2.4, the terminal recycling machine judges whether the bar code gun successfully scans and obtains the bar code data of the packing object in the step 2.2; if the code scanning is successful, executing the step 2.5-the step 2.6; if the code scanning fails, executing step 2.7;
step 2.5, the terminal recycling machine further judges whether the obtained packaged object bar code data exists in the packaged object shape feature configuration file, if not, the packaged object bar code data obtained in the step 2.4 and the packaged object feature vector space obtained in the step 2.3 are uploaded to the network service platform, and the network service platform updates the packaged object shape feature configuration file; if so, performing step 2.6;
step 2.6, the terminal recycling machine searches the configuration file of the shape characteristics of the packing object to obtain the shape characteristic data of the packing object corresponding to the bar code data of the packing object obtained in the step 2.4; the shape characteristic data of the packing object is the mean value and the normal variation range of the characteristic of the packing object;
then, the terminal recycling machine judges whether the characteristic vector space of the packing object obtained in the step 2.3 is matched with the searched shape characteristic data of the packing object, and if the matching is successful, the packing object is recycled; if the matching is unsuccessful, rejecting to recycle the wrappage;
step 2.7, the terminal recovery machine extracts a plurality of items of packing object shape characteristic data with the highest historical packing object throwing frequency from the packing object shape characteristic configuration file;
then, the characteristic vector space of the packing materials obtained in the step 2.3 is sequentially matched with the shape characteristic data of each packing material, and if the matching is successful, the packing materials are recycled; if the matching is not successful, the package is rejected for recycling.
Referring to fig. 1, the present invention further provides a recycling machine packing material shape recognition system based on infrared grating detection, including a network service platform, a recycling machine terminal module and an infrared grating collection system;
the network service platform comprises: the system comprises a platform network communication module, a packing shape feature configuration file generation module, a recycling machine configuration module and a platform alarm module;
the platform network communication module is used for communication between the network service platform and the recycling machine terminal module, and receiving and storing the packaging object bar code data uploaded by the recycling machine terminal module and the corresponding N pieces of infrared grating data; in the invention, the recovery machine terminal module communicates with the network service platform through the network, and when the network is unstable and cannot upload the message successfully, the recovery machine terminal module caches the message until the network is unobstructed and then uploads the message again. Meanwhile, in consideration of network security, the message of the recovery machine terminal module and the network service platform is subjected to an original encryption mode, so that not only can the network security be improved, but also the message content is reduced, and the size of a transmission data packet of the message is greatly reduced.
The packaging shape feature configuration file is used for carrying out feature extraction on the N pieces of infrared grating data which are received by the platform network communication module and correspond to the packaging bar code data to obtain a packaging shape feature configuration file;
the recycling machine configuration module is used for configuring relevant configuration information of a recycling machine terminal; the relevant configuration information comprises a recycling machine terminal number, longitude and latitude, an address and a configuration file;
the platform alarm module is used for processing and analyzing alarm information of the recovery machine terminal;
the recycling machine terminal module includes: the system comprises an infrared grating state detection module, a packing object identification module, a packing object throwing rebate module, a terminal alarm detection module, a terminal network communication module and a terminal maintenance management module;
the infrared grating state detection module is used for detecting whether the state of the infrared grating is normal or not;
the package identification module is used for distinguishing packages thrown into the recycling machine;
the package throwing rebate module is used for recording the number of packages thrown by a user and specific package parameter information, and uploading the package parameter information to the network service platform to realize package throwing rebate of the user;
the terminal alarm detection module is used for detecting the current running state of the recovery machine and reporting to the network service platform in an alarm mode when abnormality or hidden danger is found;
the terminal network communication module is used for communicating with the network service platform and uploading the infrared grating data acquired by the infrared grating to the network service platform; receiving a configuration file issued by the network service platform;
the terminal maintenance management module is used for realizing the checking of a maintainer on the current state of the recovery machine and the control function of the recovery machine;
the infrared grating acquisition system comprises: the system comprises a direct current motor, a belt transmission device, an infrared grating acquisition module and an infrared grating communication module;
the direct current motor and the belt transmission device are used for controlling the length and the stability of data acquired by the infrared grating acquisition module;
the infrared grating acquisition module can realize self-checking and resetting of a power-on environment and acquire data of the size of a grating shielding object in an interruption mode;
and the infrared grating communication module is used for realizing data communication between the infrared grating acquisition module and the recovery machine terminal module.
Therefore, in the invention, the recycling machine identifies the bar code on the packing object, detects the shape and the size of the packing object by an infrared grating detection principle, represents the characteristics of one packing object by utilizing a multidimensional vector space, and determines the stable point and the receiving domain of the packing object in the characteristic space according to a statistical principle so as to be used as a basis for judging whether the packing object can be recycled. When the bar code is not scanned or lost, the bar code of the package in the history package throwing record can be fully scanned and matched according to the occurrence frequency of the bar code of the package in the history package throwing record, so that the package throwing experience of a user and the positivity of the package throwing are improved.
The method and the system for identifying the shape of the packaging material of the recovery machine based on the infrared grating detection have the following advantages:
(1) the infrared grating device is adopted to detect the shape and size of the packaged object, so that the sensitivity and the precision are high, and the accuracy of feature extraction can be ensured; in addition, the infrared grating device is not easily influenced by illumination factors, and has high stability, long service life and low maintenance cost.
(2) The feature extraction algorithm has stronger robustness and fault-tolerant capability on noise, has small calculated amount, and can improve the accuracy of identifying the shape and the size of a packaged object and ensure the real-time property; obtaining a configuration file of the shape characteristics of the packaged object according to a statistical principle, eliminating abnormal data which are not in accordance with the characteristics of the bar codes, and eliminating errors generated by a motor and a conveyor belt device;
(3) the bar code is lost or not scanned, and the shape characteristic of the bar code with high package feeding history frequency is matched with the calculated infrared grating data characteristic, so that the package feeding experience of a user is enhanced, and the enthusiasm of the user for feeding packages is improved;
(4) the environmental pollution and resource waste caused by indiscriminate discarding of the packing materials are reduced, the carbon emission is reduced, and the difficulty and cost of subsequent treatment of the packing materials are greatly reduced.
The foregoing is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, various modifications and improvements can be made without departing from the principle of the present invention, and such modifications and improvements should also be considered within the scope of the present invention.
Claims (3)
1. A recycling machine packaging material shape identification method based on infrared grating detection is characterized by comprising the following steps:
step 1, generating a configuration file of the shape and the characteristics of a packaged object, and the steps are as follows:
step 1.1, configuring a bar code gun and an infrared grating acquisition device on a terminal recycling machine; the terminal recycling machine scans the bar code data of the packing materials by adopting the bar code gun; meanwhile, the terminal recycling machine adopts the infrared grating collecting device to collect N pieces of infrared grating data of the packing materials; wherein N is a natural number; the N pieces of infrared grating data are size data of N different parts of the packing material;
step 1.2, the terminal recycling machine uploads the bar code data of the packaged object and the corresponding N pieces of infrared grating data to a network service platform;
step 1.3, the network service platform performs feature extraction on the N pieces of infrared grating data corresponding to the packaging object bar code data, and maps each piece of infrared grating data to a feature vector of a packaging object feature vector space, so as to obtain a packaging object feature vector space consisting of N feature vectors;
step 1.4, rejecting abnormal feature vectors deviating from the average value to be too large according to a statistical principle for the space of the feature vectors of the packing objects, calculating the average value and the standard deviation of the feature values of the normal feature vectors according to attributes, and obtaining the average value and the normal variation range of the features of the packing objects so as to obtain the shape feature data of the packing objects corresponding to the bar code data of the packing objects; wherein the shape characteristic data of the packing object is the mean value and the normal variation range of the characteristics of the packing object;
step 1.5, continuously extracting the characteristics of the next packing material by adopting the mode of the step 1.1 to the step 1.4, and continuously circulating in such a way to obtain a plurality of packing material bar code data, wherein each packing material bar code data corresponds to the shape characteristic data of the packing material;
storing the bar code data of each packing object and the corresponding shape characteristic data of the packing object into a configuration file, thereby obtaining the configuration file of the shape characteristic of the packing object;
step 2, the actual recognition process of the shape of the packing object comprises the following steps:
step 2.1, the terminal recycling machine receives the configuration file of the shape and the characteristics of the packaged objects issued by the network service platform and loads the configuration file of the shape and the characteristics of the packaged objects into a memory;
step 2.2, when the terminal recycling machine is put into a package, the terminal recycling machine scans the package bar code data of the package by adopting the bar code gun; meanwhile, the terminal recycling machine adopts the infrared grating collecting device to collect N pieces of infrared grating data of the packing materials;
step 2.3, the terminal recycling machine performs feature extraction on the N pieces of infrared grating data acquired in the step 2.2 to obtain a packaging object feature vector space consisting of N feature vectors;
step 2.4, the terminal recycling machine judges whether the bar code gun successfully scans and obtains the bar code data of the packing object in the step 2.2; if the code scanning is successful, executing the step 2.5-the step 2.6; if the code scanning fails, executing step 2.7;
step 2.5, the terminal recycling machine further judges whether the obtained packaged object bar code data exists in the packaged object shape feature configuration file, if not, the packaged object bar code data obtained in the step 2.4 and the packaged object feature vector space obtained in the step 2.3 are uploaded to the network service platform, and the network service platform updates the packaged object shape feature configuration file; if so, performing step 2.6;
step 2.6, the terminal recycling machine searches the configuration file of the shape characteristics of the packing object to obtain the shape characteristic data of the packing object corresponding to the bar code data of the packing object obtained in the step 2.4; the shape characteristic data of the packing object is the mean value and the normal variation range of the characteristic of the packing object;
then, the terminal recycling machine judges whether the characteristic vector space of the packing object obtained in the step 2.3 is matched with the searched shape characteristic data of the packing object, and if the matching is successful, the packing object is recycled; if the matching is unsuccessful, rejecting to recycle the wrappage;
step 2.7, the terminal recovery machine extracts a plurality of items of packing object shape characteristic data with the highest historical packing object throwing frequency from the packing object shape characteristic configuration file;
then, the characteristic vector space of the packing materials obtained in the step 2.3 is sequentially matched with the shape characteristic data of each packing material, and if the matching is successful, the packing materials are recycled; if the matching is unsuccessful, rejecting to recycle the wrappage;
more specifically, step 1.1 to step 1.4 include:
1) extracting N pieces of infrared grating data corresponding to the barcode, and converting the N pieces of infrared grating data into N feature vectors in a vector space consisting of a bottle cap, a bottle head, a bottle tail, a bottle body and volume features; the algorithm for extracting the shape characteristics of the packaged objects comprises the following specific steps:
1.1) eliminating distortion data points; storing infrared grating data in data, subtracting adjacent values in the data array to obtain an absolute value, if the absolute value is greater than 10cm, then a distortion point exists at the position of the adjacent point in the array with the absolute value greater than 10cm, if the left side and the right side of the point are continuously changed and the average values of the left side and the right side are approximately equal, replacing the value of the point by the average value of the left side and the right side, otherwise, selecting the point with the minimum difference between the average value of the left side and the right side and the point by using a nearest neighbor rule to replace the value of the point;
1.2) bottle cap characteristics; the bottle cap is structurally characterized in that the difference between the outer ring of the bottle cap and the bottle neck is detected, and the difference change can be detected by adopting a high-precision grating; storing the bottle cap data in cap _ data [ ], wherein a sudden decrease change occurs from the ith point, and an increasing change rule occurs after the ith point; under the condition that the grating acquisition frequency and the belt transmission speed are fixed, averaging the first m of cap _ data to obtain the width of the bottle cap;
1.3) bottle head characteristics; the bottle head is characterized in that after the bottle cap is covered, an increasing change rule appears; the raster data generally presents a continuously increasing rule, and the change rate of the raster data is represented by the ratio of the mean value of the same points of the front section and the back section;
1.4) features of the bottle tail; the bottle tail is characterized in that the width of the bottle tail changes with different bottles, and the average value of a certain length of the bottle tail is calculated as the width of the bottle tail;
1.5) body characteristics; the width of the bottle body reflects the width of the whole bottle, and the average value of the widths from the bottle head to the bottle tail is calculated and used as the width of the bottle body;
1.6) volume characteristics; the volume characteristics are used as a rebate basis, and the key is to accurately calculate the volume of the bottle; the hatched portion indicates the section S along the beltxoy(ii) a The volume calculation formula is:
Diwidth of the wrapper in the ith cut plane, △ ziTiming the length between the cuts, grating acquisition frequency and conveyor speed, △ ziThe data is approximately constant, and n is the length of the data acquired by the grating;
2) according to the statistical principle, calculating the mean value and the standard deviation of different attribute characteristics of the N characteristic vectors according to the attribute categories, and removing the characteristic vectors which deviate from the mean value to be too large from the N characteristic vectors by using a 3 sigma rule to obtain M vectors which are corresponding to the bar codes and can truly describe the characteristics of the bottles;
3) and solving and counting the mean value and the standard deviation of each characteristic attribute of the M characteristic vectors according to the attribute types to obtain a stable point and an acceptance domain of the bottle in a vector space, and taking the stable point and the acceptance domain of each attribute characteristic as the characteristic vectors in the bottle shape library so as to judge whether the characteristic vector of one bottle conforms to the corresponding vector characteristic in the bottle shape library.
2. The method for identifying the shape of the packaging of the recycling machine based on the infrared grating detection is characterized in that in the step 1.1, the size data of the N different parts comprises the following steps: bottle cap size data, bottle head size data, bottle tail size data, bottle body size data and volume characteristic data.
3. A recoverer packaging material shape recognition system based on infrared grating detection is characterized by comprising a network service platform, a recoverer terminal module and an infrared grating acquisition system;
the network service platform comprises: the system comprises a platform network communication module, a packing shape feature configuration file generation module, a recycling machine configuration module and a platform alarm module;
the platform network communication module is used for communication between the network service platform and the recycling machine terminal module, and receiving and storing the packaging object bar code data uploaded by the recycling machine terminal module and the corresponding N pieces of infrared grating data;
the packaging shape feature configuration file is used for carrying out feature extraction on the N pieces of infrared grating data which are received by the platform network communication module and correspond to the packaging bar code data to obtain a packaging shape feature configuration file;
the package shape feature configuration file is specifically used for:
1) extracting N pieces of infrared grating data corresponding to the barcode, and converting the N pieces of infrared grating data into N feature vectors in a vector space consisting of a bottle cap, a bottle head, a bottle tail, a bottle body and volume features; the algorithm for extracting the shape characteristics of the packaged objects comprises the following specific steps:
1.1) eliminating distortion data points; storing infrared grating data in data, subtracting adjacent values in the data array to obtain an absolute value, if the absolute value is greater than 10cm, then a distortion point exists at the position of the adjacent point in the array with the absolute value greater than 10cm, if the left side and the right side of the point are continuously changed and the average values of the left side and the right side are approximately equal, replacing the value of the point by the average value of the left side and the right side, otherwise, selecting the point with the minimum difference between the average value of the left side and the right side and the point by using a nearest neighbor rule to replace the value of the point;
1.2) bottle cap characteristics; the bottle cap is structurally characterized in that the difference between the outer ring of the bottle cap and the bottle neck is detected, and the difference change can be detected by adopting a high-precision grating; storing the bottle cap data in cap _ data [ ], wherein a sudden decrease change occurs from the ith point, and an increasing change rule occurs after the ith point; under the condition that the grating acquisition frequency and the belt transmission speed are fixed, averaging the first m of cap _ data to obtain the width of the bottle cap;
1.3) bottle head characteristics; the bottle head is characterized in that after the bottle cap is covered, an increasing change rule appears; the raster data generally presents a continuously increasing rule, and the change rate of the raster data is represented by the ratio of the mean value of the same points of the front section and the back section;
1.4) features of the bottle tail; the bottle tail is characterized in that the width of the bottle tail changes with different bottles, and the average value of a certain length of the bottle tail is calculated as the width of the bottle tail;
1.5) body characteristics; the width of the bottle body reflects the width of the whole bottle, and the average value of the widths from the bottle head to the bottle tail is calculated and used as the width of the bottle body;
1.6) volume characteristics; the volume characteristics are used as a rebate basis, and the key is to accurately calculate the volume of the bottle; the hatched portion indicates the section S along the beltxoy(ii) a The volume calculation formula is:
Diwidth of the wrapper in the ith cut plane, △ ziTiming the length between the cuts, grating acquisition frequency and conveyor speed, △ ziThe data is approximately constant, and n is the length of the data acquired by the grating;
2) according to the statistical principle, calculating the mean value and the standard deviation of different attribute characteristics of the N characteristic vectors according to the attribute categories, and removing the characteristic vectors which deviate from the mean value to be too large from the N characteristic vectors by using a 3 sigma rule to obtain M vectors which are corresponding to the bar codes and can truly describe the characteristics of the bottles;
3) solving and counting the mean value and standard deviation of each characteristic attribute of the M characteristic vectors according to the attribute types to obtain a stable point and an acceptance domain of the bottle in a vector space, and taking the stable point and the acceptance domain of each attribute characteristic as the characteristic vectors in the bottle shape library so as to judge whether the characteristic vector of one bottle conforms to the corresponding vector characteristic in the bottle shape library;
the recycling machine configuration module is used for configuring relevant configuration information of a recycling machine terminal; the relevant configuration information comprises a recycling machine terminal number, longitude and latitude, an address and a configuration file;
the platform alarm module is used for processing and analyzing alarm information of the recovery machine terminal;
the recycling machine terminal module includes: the system comprises an infrared grating state detection module, a packing object identification module, a packing object throwing rebate module, a terminal alarm detection module, a terminal network communication module and a terminal maintenance management module;
the infrared grating state detection module is used for detecting whether the state of the infrared grating is normal or not;
the package identification module is used for distinguishing packages thrown into the recycling machine;
the package throwing rebate module is used for recording the number of packages thrown by a user and specific package parameter information, and uploading the package parameter information to the network service platform to realize package throwing rebate of the user;
the terminal alarm detection module is used for detecting the current running state of the recovery machine and reporting to the network service platform in an alarm mode when abnormality or hidden danger is found;
the terminal network communication module is used for communicating with the network service platform and uploading the infrared grating data acquired by the infrared grating to the network service platform; receiving a configuration file issued by the network service platform;
the terminal maintenance management module is used for realizing the checking of a maintainer on the current state of the recovery machine and the control function of the recovery machine;
the infrared grating acquisition system comprises: the system comprises a direct current motor, a belt transmission device, an infrared grating acquisition module and an infrared grating communication module;
the direct current motor and the belt transmission device are used for controlling the length and the stability of data acquired by the infrared grating acquisition module;
the infrared grating acquisition module can realize self-checking and resetting of a power-on environment and acquire data of the size of a grating shielding object in an interruption mode;
and the infrared grating communication module is used for realizing data communication between the infrared grating acquisition module and the recovery machine terminal module.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710517274.1A CN107514978B (en) | 2017-06-29 | 2017-06-29 | Recovery machine packing material shape identification method and system based on infrared grating detection |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710517274.1A CN107514978B (en) | 2017-06-29 | 2017-06-29 | Recovery machine packing material shape identification method and system based on infrared grating detection |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107514978A CN107514978A (en) | 2017-12-26 |
| CN107514978B true CN107514978B (en) | 2020-07-14 |
Family
ID=60722019
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710517274.1A Active CN107514978B (en) | 2017-06-29 | 2017-06-29 | Recovery machine packing material shape identification method and system based on infrared grating detection |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107514978B (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109230061A (en) * | 2018-08-20 | 2019-01-18 | 叶志徐 | A kind of 3-D scanning recycling machine and its recovery method |
| AU2019404076A1 (en) | 2018-12-19 | 2021-07-15 | Ecoatm, Llc | Systems and methods for vending and/or purchasing mobile phones and other electronic devices |
| AU2020222971A1 (en) | 2019-02-12 | 2021-09-23 | Ecoatm, Llc | Connector carrier for electronic device kiosk |
| JP2021530793A (en) | 2019-02-18 | 2021-11-11 | エコエーティーエム， エルエルシー | Neural networks based on physical state assessment of electronic devices, and associated systems and methods |
| USD1010271S1 (en) | 2020-02-12 | 2024-01-02 | Ecoatm, Llc | Kiosk for processing electronic devices |
| CN111931557B (en) * | 2020-06-19 | 2024-05-10 | 广州图匠数据科技有限公司 | Method and device for identifying specification of bottled drink, terminal equipment and readable storage medium |
| US11922467B2 (en) | 2020-08-17 | 2024-03-05 | ecoATM, Inc. | Evaluating an electronic device using optical character recognition |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN202025377U (en) * | 2011-04-14 | 2011-11-02 | 威海新北洋数码科技股份有限公司 | Bottle recognition device and automatic bottle recovery equipment |
| CN202548962U (en) * | 2012-05-10 | 2012-11-21 | 吴方 | Recycling device based on image recognition device |
| CN103217127A (en) * | 2013-05-03 | 2013-07-24 | 郑州恒科实业有限公司 | Infrared light curtain machine for contour recognition and article contour recognition system and method thereof |
| CN105354920A (en) * | 2015-12-03 | 2016-02-24 | 杭州步展科技有限公司 | Intelligent drink bottle recycling machine |
| CN105959408A (en) * | 2016-06-27 | 2016-09-21 | 北京盈创高科新技术发展有限公司 | Real time recycling rebating method and system for standard packaging material recycling machine |
| CN106846621A (en) * | 2017-02-20 | 2017-06-13 | 湖南大学 | " internet+" Packaging Bottle intelligence recovery system and method |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH09210651A (en) * | 1996-01-29 | 1997-08-12 | Kirin Techno Syst:Kk | Identification method for recycling bottle |
| CN204463286U (en) * | 2015-01-26 | 2015-07-08 | 东莞职业技术学院 | Beverage bottle recycling machine |
| CN106097574B (en) * | 2016-06-27 | 2018-08-31 | 北京盈创高科新技术发展有限公司 | Container based on Internet of Things pattern intelligently recycles self-aided terminal and recovery method |
-
2017
- 2017-06-29 CN CN201710517274.1A patent/CN107514978B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN202025377U (en) * | 2011-04-14 | 2011-11-02 | 威海新北洋数码科技股份有限公司 | Bottle recognition device and automatic bottle recovery equipment |
| CN202548962U (en) * | 2012-05-10 | 2012-11-21 | 吴方 | Recycling device based on image recognition device |
| CN103217127A (en) * | 2013-05-03 | 2013-07-24 | 郑州恒科实业有限公司 | Infrared light curtain machine for contour recognition and article contour recognition system and method thereof |
| CN105354920A (en) * | 2015-12-03 | 2016-02-24 | 杭州步展科技有限公司 | Intelligent drink bottle recycling machine |
| CN105959408A (en) * | 2016-06-27 | 2016-09-21 | 北京盈创高科新技术发展有限公司 | Real time recycling rebating method and system for standard packaging material recycling machine |
| CN106846621A (en) * | 2017-02-20 | 2017-06-13 | 湖南大学 | " internet+" Packaging Bottle intelligence recovery system and method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107514978A (en) | 2017-12-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107514978B (en) | Recovery machine packing material shape identification method and system based on infrared grating detection | |
| CN104303193B (en) | Target classification based on cluster | |
| US9064187B2 (en) | Method and system for item identification | |
| WO2014198315A1 (en) | Image based object classification | |
| CN105205431A (en) | Two-dimensional code automatic correction system and method in packaging process | |
| CN109766891B (en) | Method for acquiring equipment facility information and computer readable storage medium | |
| KR101978109B1 (en) | Methods for determining if a mark is genuine, | |
| CN101976350A (en) | Grain storage pest detection and identification method based on video analytics and system thereof | |
| CN111573039A (en) | Intelligent garbage classification processing system based on big data | |
| CN110991570B (en) | Article monitoring method based on ultrahigh frequency RFID and picture identification | |
| JP6487565B2 (en) | Mail image pairing using gradient field singularity descriptors | |
| CN113723176B (en) | Target object determination method and device, storage medium and electronic device | |
| CN106600287A (en) | Intelligent anti-counterfeiting early warning method and device | |
| CN110502605A (en) | Electric power asset LCC cost accumulation system based on artificial intelligence technology | |
| CN116976375A (en) | Method and device for detecting cigarette packaging box in code scanning and verification scene | |
| CN112613362A (en) | Article mark identification system based on Internet of things | |
| CN116503848A (en) | Intelligent license plate recognition method, device, equipment and storage medium | |
| CN115860605A (en) | Warehouse logistics detection system based on visual identification | |
| CN109871723B (en) | Image fusion information processing method and system | |
| CN109815764B (en) | Method and system for reading machine-readable information in image | |
| CN111918231A (en) | Baggage mobile intelligent network management method | |
| CN116206093B (en) | Electric meter data acquisition method and system based on bitmap and readable storage medium | |
| CN114038113A (en) | System and method for identifying multidimensional shapes of packaging materials of recycling machine based on infrared grating | |
| Breuel | Recognition by Adaptive Subdivision of Transformation Space: practical experiences and comparison with the Hough transform | |
| CN114743082A (en) | Cigarette finished product cigarette box characteristic identification method based on machine vision |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 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 |