CN114378002A - Hericium erinaceus grading system and grading method based on machine vision - Google Patents
Hericium erinaceus grading system and grading method based on machine vision Download PDFInfo
- Publication number
- CN114378002A CN114378002A CN202210222352.6A CN202210222352A CN114378002A CN 114378002 A CN114378002 A CN 114378002A CN 202210222352 A CN202210222352 A CN 202210222352A CN 114378002 A CN114378002 A CN 114378002A
- Authority
- CN
- China
- Prior art keywords
- hericium erinaceus
- grading
- grade
- quality
- image
- 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
- 240000000588 Hericium erinaceus Species 0.000 title claims abstract description 245
- 235000007328 Hericium erinaceus Nutrition 0.000 title claims abstract description 245
- 238000000034 method Methods 0.000 title claims abstract description 47
- 238000001514 detection method Methods 0.000 claims abstract description 21
- 238000012545 processing Methods 0.000 claims abstract description 19
- 230000007547 defect Effects 0.000 claims abstract description 14
- 238000000605 extraction Methods 0.000 claims abstract description 8
- 238000004458 analytical method Methods 0.000 claims abstract description 7
- 230000005540 biological transmission Effects 0.000 claims description 12
- 239000002245 particle Substances 0.000 claims description 8
- 230000000007 visual effect Effects 0.000 claims description 8
- 235000020985 whole grains Nutrition 0.000 claims description 8
- 238000003708 edge detection Methods 0.000 claims description 7
- 230000007246 mechanism Effects 0.000 claims description 7
- 238000003062 neural network model Methods 0.000 claims description 7
- 238000007781 pre-processing Methods 0.000 claims description 5
- 230000011218 segmentation Effects 0.000 claims description 5
- 230000006870 function Effects 0.000 claims description 4
- 238000011160 research Methods 0.000 claims description 4
- 238000012549 training Methods 0.000 claims description 4
- 238000004519 manufacturing process Methods 0.000 description 10
- 239000000463 material Substances 0.000 description 10
- 235000001674 Agaricus brunnescens Nutrition 0.000 description 8
- 241000289669 Erinaceus europaeus Species 0.000 description 8
- 241000577951 Hydnum Species 0.000 description 8
- 239000004814 polyurethane Substances 0.000 description 8
- 238000010586 diagram Methods 0.000 description 5
- 229920003023 plastic Polymers 0.000 description 5
- 239000004033 plastic Substances 0.000 description 4
- 229920002635 polyurethane Polymers 0.000 description 4
- 238000012546 transfer Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 235000013305 food Nutrition 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 238000001035 drying Methods 0.000 description 2
- 230000002708 enhancing effect Effects 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 230000001965 increasing effect Effects 0.000 description 2
- 238000009434 installation Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 229910001220 stainless steel Inorganic materials 0.000 description 1
- 239000010935 stainless steel Substances 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
- B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
- B07C5/00—Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
- B07C5/34—Sorting according to other particular properties
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
- B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
- B07C5/00—Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
- B07C5/02—Measures preceding sorting, e.g. arranging articles in a stream orientating
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
- B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
- B07C5/00—Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
- B07C5/04—Sorting according to size
- B07C5/10—Sorting according to size measured by light-responsive means
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
- B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
- B07C5/00—Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
- B07C5/34—Sorting according to other particular properties
- B07C5/342—Sorting according to other particular properties according to optical properties, e.g. colour
- B07C5/3422—Sorting according to other particular properties according to optical properties, e.g. colour using video scanning devices, e.g. TV-cameras
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
- B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
- B07C5/00—Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
- B07C5/36—Sorting apparatus characterised by the means used for distribution
- B07C5/361—Processing or control devices therefor, e.g. escort memory
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
- B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
- B07C5/00—Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
- B07C5/36—Sorting apparatus characterised by the means used for distribution
- B07C5/38—Collecting or arranging articles in groups
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
- B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
- B07C2501/00—Sorting according to a characteristic or feature of the articles or material to be sorted
- B07C2501/0081—Sorting of food items
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Sorting Of Articles (AREA)
Abstract
The invention discloses a hericium erinaceus grading system and a grading method based on machine vision, belongs to the technical field of agricultural product processing and intelligent detection nondestructive grading, and particularly relates to a hericium erinaceus grading system and a grading method based on machine vision. The system comprises a hardware module controlled by an industrial control integrated machine, a characteristic parameter extraction and analysis module and a quality grading module. Wherein the hardware module mainly includes: the device comprises a hericium erinaceus belt type conveying and arranging device, a CCD industrial camera, an annular light source, a rotatable circular grading conveying device, a corresponding grading conveying channel, an industrial personal computer, an anti-slip baffle, a fixed support, an image acquisition card and corresponding connecting and assembling basic parts. The grading method judges the grade of the hericium erinaceus by processing images of the hericium erinaceus and extracting characteristic parameters, can objectively, accurately and efficiently automatically grade the quality of the hericium erinaceus, makes up for the defect of manual grading, and realizes online nondestructive rapid detection and grading of the grade of the hericium erinaceus.
Description
Technical Field
The invention relates to detection and judgment of hericium erinaceus quality grade, belongs to the technical field of agricultural product processing and intelligent detection nondestructive grading, and particularly relates to a hericium erinaceus grading system and method based on machine vision.
Background
At the present stage, the automation and intelligence degree of the production and processing process of the hericium erinaceus are low, and the grading process of the hericium erinaceus is mainly completed manually. The manual grading mode is large in workload and high in strength, and the repeated same work is easy to fatigue, so that the efficiency is low, the error rate is high, and the production cost of the hericium erinaceus is increased.
The hericium erinaceus needs to be classified according to different qualities. Generally divided into first-level, second-level, third-level, fourth-level and off-level. At the present stage, the automation and intelligence degree of the production and processing process of the hericium erinaceus is lower, and the quality grading process of the hericium erinaceus after the drying and processing is mainly completed manually. The mode has the advantages of huge workload, high strength, repeated large amount of same work, waste of a large amount of manpower and material resources and increase of the production cost of the hericium erinaceus; in the whole production process, the labor is easy to fatigue, so that the efficiency is low, the error rate is high, the subjectivity is strong, the quality grade standard is disordered, the market price of the hericium erinaceus is high, and the economic benefit is low. In recent years, with the annual increase in the scale of growing hericium erinaceus and the increasing activity of market trading, the negative effects of these problems have become more prominent.
Therefore, it is necessary to deeply research a grade grading method and technology of hericium erinaceus so as to develop an objective, accurate and efficient automatic grading system and grading method for the quality of the hericium erinaceus, so that the hericium erinaceus is graded according to external characteristics such as color, shape, integrity and the like of the hericium erinaceus, the defect of manual grading is overcome, and online nondestructive rapid detection and grading of the grade of the hericium erinaceus is realized.
Disclosure of Invention
The technical problems to be solved by the invention are as follows: the invention aims to solve the problem that real-time online detection and classification cannot be realized in the existing hericium erinaceus processing process, improve the objectivity, accuracy and working efficiency of hericium erinaceus quality detection, and realize real-time online quality nondestructive detection and automatic classification in the hericium erinaceus processing process.
In order to solve the technical problems, the invention provides the following technical scheme:
a hericium erinaceus automatic grading system based on machine vision comprises a hardware module, a characteristic parameter extraction and analysis module and a quality grading module, wherein the hardware module, the characteristic parameter extraction and analysis module and the quality grading module are controlled by an industrial control integrated machine; wherein, the hardware module includes: the device comprises a hericium erinaceus belt type conveying and arranging device, a CCD industrial camera, an annular light source, an image acquisition card, a rotatable circular grading and conveying device, a corresponding grading and conveying channel, an industrial personal computer, an anti-slip baffle, a fixed support and a corresponding connecting and assembling basic part; the characteristic parameter extraction and analysis module collects preset hericium erinaceus appearance characteristic parameters through a hardware module, and the hericium erinaceus appearance characteristic parameters are as follows: the color C of the hericium erinaceus, the shape S of the hericium erinaceus, the integrity I of the hericium erinaceus and the diameter D of the hericium erinaceus; the quality grading module carries out modeling prediction on the acquired hericium erinaceus appearance characteristic parameters through an RGB model, a circularity model, a minimum circumcircle method and a BP neural network model, and then grades of the hericium erinaceus are divided into five grades including a special grade, a first grade, a second grade, a third grade and a grade.
A hericium erinaceus grading method adopts the automatic grading system for grading hericium erinaceus based on machine vision, and the grading method comprises the following steps: the hericium erinaceus are arranged before detection through the belt type conveying and arranging device, transmission and detection of single hericium erinaceus can be achieved after arrangement, when the hericium erinaceus pass through the CCD industrial camera and the lower portion of the annular light source through the conveying belt, collection of images of the hericium erinaceus is guaranteed to be completed under the condition that the light source is sufficient, corresponding hericium erinaceus appearance characteristic parameters are obtained, then modeling prediction is conducted through the grading module according to the hericium erinaceus appearance characteristic parameters, and quality grades of the hericium erinaceus are automatically judged; the judged grade information and the real-time position information of the hericium erinaceus are transmitted to the industrial control all-in-one machine, the industrial control all-in-one machine controls the rotary circular grading transmission device to rotate according to the control of the corresponding grade of the hericium erinaceus, the hericium erinaceus are transmitted to the corresponding grade transmission channel, and the whole grading process is completed.
Preferably, the specific steps are as follows:
(1) manually grading and recording the quality grade of the hericium erinaceus: the hericium erinaceus is rated by manual observation of a research or a worker who is trained and has rating experience, and the rating standard is as follows: the first level, the second level, the third level, the fourth level and 5 levels outside the levels are respectively recorded by numerical values 1, 2, 3, 4 and 5;
(2) collecting images of the hericium erinaceus: after the hericium erinaceus pass through the belt type conveying and arranging device, the hericium erinaceus sequentially and singly enter an image collecting area, the hericium erinaceus are conveyed to the position under the CCD industrial camera and the annular light source through the conveying belt, the camera collects images of the hericium erinaceus, and the collected images are conveyed to an automatic grading stage of the hericium erinaceus device through a connecting line and an image collecting card;
(3) extracting the quality characteristic parameters of the hericium erinaceus: the hericium erinaceus quality automatic grading software is represented by 4 quality characteristic parameters according to the collected image characteristics;
(4) establishing a hericium erinaceus quality grade database: repeating the steps (1) to (3), and establishing a hericium erinaceus quality grade database, wherein each record of the database consists of 4 characteristic parameters in the step (3) and the hericium erinaceus quality grade in the step (1);
(5) establishing a hericium erinaceus quality grading model: taking each record in a hericium erinaceus quality grade database as a training set, taking 4 characteristic parameters of each record as input, taking hericium erinaceus quality grade as output, and establishing a hericium erinaceus quality grading model according to a modeling method of an RGB model, a circularity model, a minimum circumcircle and a BP neural network model;
(6) automatically grading the quality of the hericium erinaceus: placing hericium erinaceus which are not subjected to manual grading on a conveyor belt of a belt conveyor, collecting image information of the hericium erinaceus to be detected under a CCD industrial camera, transmitting the image information to a hericium erinaceus automatic grading module of the industrial control all-in-one machine, finishing image processing by the hericium erinaceus automatic grading module within 10ms, extracting 4 hericium erinaceus appearance characteristic parameters from the images, and comparing the extracted hericium erinaceus appearance characteristic parameters with the hericium erinaceus quality automatic grading model established in the step (5) to obtain the quality of the hericium erinaceus;
(7) after the hericium erinaceus passes through the visual detection area, the hericium erinaceus enters the classification area, the classification area is provided with a first level, a second level, a third level and five classification conveying channels outside the levels along the conveying belt direction, the executing mechanism adopts a rotatable circular classification conveying device, the industrial control all-in-one machine carries out real-time control on the steering of the conveying belt through the level information output by visual system recognition and the current position information of the hericium erinaceus, the circular rotatable conveying device corresponding to the levels steers, and the hericium erinaceus is conveyed to the classification conveying channels corresponding to the levels.
Preferably, the rating criteria are as follows:
if the grading model is characterized by light yellow, the whole particle is circular without any defect, the diameter is more than 5-6 cm, the grade is 1; if the grading model is characterized by light yellow, the whole particle is circular without any defect, the diameter is 4-5 cm, the grade is two, and the grade is 2;
if the grading model is characterized by deep yellow, the whole grain is approximately round without any defect, the diameter is 3-4 cm, the grade is 3; if the grading model is characterized by dark yellow, the whole grain is approximately round and slightly incomplete, the diameter is 1-3 cm, the grade is marked as 4; if the grading model features blackened or whitish, obvious, malformed, incomplete, the grade is outside the grade, and the grade is marked as 5.
Preferably, the method for extracting 4 hericium erinaceus appearance characteristic parameters from the image in the step (6) comprises the following steps:
preprocessing the acquired image by combining the image processing software and the grading rule of a grading system with a computer programming language, performing depth processing on the preprocessed image, selecting a proper threshold value, extracting color features of the image through an RGB (red, green and blue) model, and calculating an obtained histogram by using a function calcHist () function carried by Opencv to realize estimation of the color range of the hericium erinaceus; segmenting the background part of the hericium erinaceus image and the hericium erinaceus image by adopting a threshold segmentation method, acquiring whether the shape of the hericium erinaceus tends to be circular or not, and judging the integrity of the hericium erinaceus; and (4) performing edge detection on the contour region of the hericium erinaceus by adopting Canny edge detection, and calculating the diameter of the hericium erinaceus by using a minimum circumcircle method.
The invention has the following beneficial effects:
according to the invention, hericium erinaceus are orderly arranged and conveyed onto a conveying belt by adopting a material shifting mechanism and a hericium erinaceus belt type conveying and arranging device, the conveying belt pulls the individual spacing of the hericium erinaceus apart by utilizing a differential principle, and only one object to be detected is ensured to appear in the visual field range of an industrial camera, so that the hericium erinaceus can pass through an image acquisition region consisting of a CCD industrial camera, an industrial lens and an annular light source one by one to realize the acquisition of image data of the hericium erinaceus, model building and data analysis are carried out on the acquired data including relevant parameters of the characteristics of the color, the diameter, the shape, the integrity and the like of the hericium erinaceus, and the grade (first grade, second grade, third grade, fourth grade and beyond grade) of the hericium erinaceus is predicted. Realize the automatic grading of hedgehog hydnum mushroom grade, saved hiring of the rating personnel on the production line, the corresponding spending of training, standardize hedgehog hydnum mushroom quality rating standard, promote hedgehog hydnum mushroom quality simultaneously and detect objectivity, rate of accuracy and work efficiency, realize online quality nondestructive test and automatic grading in the hedgehog hydnum mushroom course of working.
Drawings
FIG. 1 is a hierarchical system workflow diagram of the present invention.
Fig. 2 is a schematic diagram of the mechanical structural framework of the grading system of the present invention.
FIG. 3 is a detailed view of the delivery staging device of the present invention.
FIG. 4 is a diagram of the image acquisition module hardware detail of the present invention.
FIG. 5 is a detailed view of the rotatable staging conveyor of the present invention.
The reference numerals in the figures illustrate: 1-a transfer carriage; 2-a material poking mechanism; 3-an image acquisition component; 4-a rotatable grading mechanism; 5, a conveyor belt; 6, an industrial control integrated machine; 7-hierarchical level channels; 8-image acquisition card.
FIG. 6 is a schematic diagram of the classification process of the present invention.
Fig. 7 is an example of an image in the image processing process of the present invention.
Detailed Description
The following examples are included to provide further detailed description of the present invention and to provide those skilled in the art with a more complete, concise, and exact understanding of the principles and spirit of the invention.
Example 1: a work flow chart of the system of the automatic grading system for the hericium erinaceus based on machine vision is shown in figure 1 and comprises a hardware module, a characteristic parameter extraction and analysis module and a quality grading module which are controlled by an industrial control all-in-one machine. Wherein the hardware module mainly includes: the device comprises a hericium erinaceus belt type conveying and arranging device, a CCD industrial camera, an annular light source, a rotatable circular grading and conveying device, a corresponding grading and conveying channel, an industrial personal computer, an anti-slip baffle, a fixed support and corresponding connecting and assembling basic parts; the characteristic parameter draws the hericium erinaceus appearance characteristic parameter that the module mainly draws and obtains by industry control all-in-one behind the hardware module control effect, totally 4, include: the color C of the hericium erinaceus, the shape S of the hericium erinaceus, the integrity I of the hericium erinaceus and the diameter D of the hericium erinaceus; the quality grading module is mainly used for modeling and predicting according to the acquired characteristic parameters through an RGB (red, green and blue) model, a circularity model and a BP (back propagation) neural network model, and grading the quality grade of the hericium erinaceus into a first grade, a second grade, a third grade, a fourth grade and five grades except the first grade.
As shown in fig. 2, when the hericium erinaceus is conveyed to the image acquisition and detection module by the conveyor belt 5, the annular light source illuminates an image acquisition area to provide sufficient light source, the CCD industrial camera acquires images of the hericium erinaceus, and the detection device performs dynamic positioning and visual detection on the hericium erinaceus during operation, which belongs to the prior art and the description is not elaborated in detail; the collected images are connected with an image collecting card 8 through a USB connecting line and are transmitted to an industrial control all-in-one machine 6, the industrial control all-in-one machine 6 finishes image processing and grade judgment, and furthermore, hericium erinaceus are transmitted to a grading transmission stage and are transmitted to a corresponding grade transmission channel 7 through a rotatable circular grading transmission device 4 to finish grading of the hericium erinaceus. The image acquisition assembly 3 consists of a CCD industrial camera, a lens and an annular light source. The CCD industrial camera selects an MV-EM120C industrial camera, the maximum resolution is 1280 multiplied by 960, and the vertical distance between a lens and a conveyor belt is fixed to be 160 mm; the annular light source is lighted by the LED lamp; the CCD industrial camera and the annular light source are fixed through the fixing support and the fixing baffle.
As shown in the attached drawing 3, the belt type conveying and arranging device for hericium erinaceus mainly aims to arrange a plurality of hericium erinaceus on a production line in order, convey the hericium erinaceus singly into a detection and classification module for classification and grade judgment, then enter a classification conveying part of a hardware device, and finally finish classification of the hericium erinaceus with corresponding grades. The whole conveying device is divided into four parts, namely a conveying part with different widths and capable of independently operating, a rotatable circular grading conveying part, a corresponding grading conveying channel part and a vertical baffle part with gradually changed width. Wherein the belt conveyer part, the corresponding grade conveying channel part and the rotatable circular grading conveying part are made of food grade white PU (polyurethane) plastic materials, the vertical baffle is made of 316 food grade stainless steel materials, the height is 60mm, and the vertical baffle is fixed above the conveyer belt in a slight contact way and does not influence the operation of the conveyer belt. The width is 300mm after its conveyer belt is acted on by vertical baffle on the belt conveyor collating unit left side (hedgehog hydnum mushroom initial direction of motion), and the width reduces to 150mm after the baffle effect, realizes the one-way single transportation of hedgehog hydnum mushroom, then reduces to 100mm again, and the hierarchical conveying stage of entering device realizes that single hedgehog hydnum mushroom is hierarchical. The conveying speed of the conveyor belt 5 and the rotating frequency of the rotatable circular grading conveyor 4 corresponding to the grading conveying channel 7 can be controlled and adjusted by the hericium erinaceus quality grading module.
As shown in fig. 4, the hardware part of the image acquisition and detection module mainly includes: an annular light source 31; a CCD industrial camera 32; a CCD industrial lens 33. The CCD industrial camera 32 selects an MV-EM120C industrial camera, the maximum resolution is 1280 multiplied by 960, the size is 35mm multiplied by 29mm, and the vertical distance between the lens and the conveyor belt is fixed to be 160 mm; the annular light source 31 is lit by an LED lamp; the fixed baffle plate is made of transparent plastic material 601, and the corresponding hardware device is mainly fixed by the fixed bracket and the fixed baffle plate. And the tasks of acquisition and input of image information data of the hericium erinaceus, extraction of characteristic parameters, judgment and output of quality grades and the like are correspondingly completed in sequence.
As shown in fig. 2, 3 and 5, when the hericium erinaceus moves to the grading conveying part, the hericium erinaceus automatically grades the quality degree of the hericium erinaceus mainly by the industrial control integrated machine 6 and corresponding automatic grading software, and the industrial control integrated machine 6 controls the rotatable circular grading conveying device 4 according to the judgment grade information and the position information of the hericium erinaceus. The rotatable circular grading conveyor 4 is composed of 8 rotatable disc small conveyors with the diameter of 100mm, and is made of food-grade white PU (polyurethane) plastic materials; rotatable circular hierarchical conveyer 4 conveys the hedgehog hydnum mushroom to corresponding grade transfer passage 7, and corresponding grade transfer passage 7 material is food level white PU (polyurethane) plastics material, and the width is 160mm, and it is fixed 350mm to correspond grade transfer passage 7 bottommost (terminal) and conveyer belt 5 vertical distance. The grading conveyor part is composed of 2 rotatable circular grading conveyors 4 and 5 grading corresponding conveying channels 7 and completes grading conveying tasks of the hericium erinaceus with corresponding grades according to grade information and position information of the hericium erinaceus output by the system.
FIG. 6 is a schematic diagram of the classification process of the present invention. And the industrial control all-in-one machine 6 controls the CCD industrial camera to acquire images of the hericium erinaceus passing through the image acquisition area and prepare for the next image processing. Secondly, preprocessing the acquired image by combining image processing software and a grading rule of a grading system with a computer programming language, wherein the preprocessing comprises graying of the image, denoising of the image, enhancing of the image and the like, and the preprocessed image needs to be subjected to deep processing; further, in order to obtain whether the shape of the hericium erinaceus tends to be circular, a threshold segmentation method is adopted to segment the background part of the hericium erinaceus image and the hericium erinaceus image, and the integrity of the hericium erinaceus is judged; on the basis, the edge detection is carried out on the outline region of the hericium erinaceus by adopting Canny edge detection, and the diameter of the hericium erinaceus is calculated by a minimum circumcircle method. And finally, carrying out grade prediction on the characteristic parameters obtained after collection and operation through an industrial personal computer, thereby automatically grading.
Fig. 7 is an image example in the image processing process of the present invention, where an image a is an acquired original image, an industrial personal computer performs preprocessing (graying of the image, denoising of the image, and enhancing of the image) on the original image to obtain an image B, analyzes the image through an RGB model to obtain a chromaticity histogram C corresponding to RGB in order to obtain color characteristics of the hericium erinaceus, and further performs threshold segmentation processing by using a threshold segmentation method in order to judge the integrity of the hericium erinaceus. In order to calculate the average diameter of the hericium erinaceus, Canny edge detection is firstly carried out to obtain an image E, then an image F is obtained by adopting a minimum circumcircle method, and the diameter of the hericium erinaceus is accurately judged. And finally, comparing the collected parameters with the judgment rule of the hericium erinaceus to predict the grade.
The method for grading the hericium erinaceus by using the machine vision-based grading system mainly comprises the following steps of:
(1) manually grading and recording the quality grade of the hericium erinaceus: the hericium erinaceus is rated by manual observation of a research or a worker who is trained and has rating experience, and the rating standard is as follows: the first grade, the second grade, the third grade, the fourth grade and 5 grades outside the grades are recorded by numerical values 1, 2, 3, 4 and 5. The criteria for determination are as follows:
if the grading model is characterized by light yellow, the whole particle is circular without any defect, the diameter is more than 5-6 cm, the grade is 1;
if the grading model is characterized by light yellow, the whole particle is circular without any defect, the diameter is 4-5 cm, the grade is two, and the grade is 2;
if the grading model is characterized by deep yellow, the whole grain is approximately round without any defect, the diameter is 3-4 cm, the grade is 3;
if the grading model is characterized by dark yellow, the whole grain is approximately round and slightly incomplete, the diameter is 1-3 cm, the grade is marked as 4;
if the grading model features blackened or whitish, obvious, malformed, incomplete, the grade is outside the grade, and the grade is marked as 5.
(2) Collecting images of the hericium erinaceus: the hericium erinaceus are conveyed to the arrangement device through the belt type conveying device, the hericium erinaceus sequentially enter the image collecting area, the conveyor belt conveys the hericium erinaceus to the CCD industrial camera and the annular light source, the industrial camera collects images of the hericium erinaceus, and the collected images are conveyed to the hericium erinaceus quality automatic grading module.
(3) Extracting the quality characteristic parameters of the hericium erinaceus: the automatic grading module for the hericium erinaceus quality is represented by 4 quality characteristic parameters according to the collected image characteristics, and the characteristics are described as follows:
the method is characterized in that: color of Hericium erinaceus C.
And (2) feature: hericium erinaceus is shaped as S.
And (3) feature: integrity of Hericium erinaceus I.
And (4) feature: average diameter D of Hericium erinaceus.
(4) Establishing a hericium erinaceus quality grade database: and (4) repeating the steps (1) to (3), and establishing a hericium erinaceus quality grade database, wherein each record of the database consists of 4 characteristic parameters in the step (3) and the hericium erinaceus quality grade in the step (1).
(5) Establishing a hericium erinaceus quality grading model: and (3) taking each record in the hericium erinaceus quality grade database as a training set, taking 4 characteristic parameters of each record as input, taking the hericium erinaceus quality grade as output, and establishing a hericium erinaceus quality grading model according to a modeling method of an RGB (red, green and blue) model, a circularity model, a minimum circumcircle and a BP (back propagation) neural network model.
(6) Automatically grading the quality of the hericium erinaceus: arranging the hericium erinaceus which are not subjected to manual grading on a conveyor belt of a belt conveyor, allowing the hericium erinaceus to be detected to pass through a CCD industrial camera and below the CCD industrial camera, collecting image information of the hericium erinaceus and transmitting the image information to an automatic grading module of the hericium erinaceus of the industrial control all-in-one machine, finishing image processing by the automatic grading module of the hericium erinaceus within 10ms, extracting 4 characteristic parameters from the images, and finally obtaining the quality of the hericium erinaceus according to the automatic grading model of the quality of the hericium erinaceus established in the step (5). Since the hierarchical model is an RGB model, a circularity model and a BP neural network model, in order to make a system make a specific judgment, it is necessary to specifically specify the grade of hericium erinaceus quality as follows:
if the grading model is characterized by light yellow, the whole particle is circular without any defect, the diameter is more than 5-6 cm, the grade is 1;
if the grading model is characterized by light yellow, the whole particle is circular without any defect, the diameter is 4-5 cm, the grade is two, and the grade is 2;
if the grading model is characterized by deep yellow, the whole grain is approximately round without any defect, the diameter is 3-4 cm, the grade is 3;
if the grading model is characterized by dark yellow, the whole grain is approximately round and slightly incomplete, the diameter is 1-3 cm, the grade is marked as 4;
if the grading model features blackened or whitish, obvious, malformed, incomplete, the grade is outside the grade, and the grade is marked as 5.
(7) After the hericium erinaceus passes through the visual detection area, the hericium erinaceus enters the classification area, the classification area is provided with a first level, a second level, a third level and five classification conveying channels outside the four levels and the grade, five-level classification of the hericium erinaceus is realized, the classification structure mainly adopts a rotatable circular classification conveying device (for better rotation, each classification hardware system consists of 8 small circular conveying mechanisms), the industrial control all-in-one machine conducts real-time control on the steering of the conveying belt through the grade information recognized and output by the visual system and the current position information of the hericium erinaceus, the circular rotatable conveying device corresponding to the grade is made to steer, and the hericium erinaceus is made to be conveyed to the classification channels of the corresponding grade. The conveying part is divided into three sections, the front section is mainly a fixed conveying part of the hericium erinaceus drying production line, the speed and the number of the hericium erinaceus are consistent with those of a normal production line, and before the conveying part is connected, the width of the conveying belt is narrowed; the middle section is kept narrowed, and then a single hericium erinaceus is transmitted in a fixed width mode to carry out image acquisition and grading detection; the back section is narrowed continuously, and in the stage of grading transmission of a mechanical structure, a single hericium erinaceus is transmitted and transmitted to the corresponding grade transmission channel through the rotatable circular grading transmission device, and grading of the hericium erinaceus in the corresponding grade is completed.
The selection of the industrial camera and the height of the installation position are obtained through repeated experiments of a large number of theoretical experiments, so that images of all levels of hericium erinaceus can be clearly acquired, and the external characteristics of the hericium erinaceus can be accurately distinguished in real time.
The annular light source is specially designed and is arranged at a position which takes the center of a lens of the industrial camera as a circle center, is in the same horizontal line with the lens of the industrial camera and surrounds the industrial camera. The LED lamp is used for lighting, the brightness is sufficient, the image acquisition process can be guaranteed, the brightness is sufficient, and the acquired hericium erinaceus picture is clear enough.
The annular rotatable grading conveying device is specially designed, in order to improve the working efficiency of the system, each 8 small circular conveying devices form a large conveying mechanism, each small circular conveying device can rotate 360 degrees and can rotate 10 times per second, the grading efficiency of the hericium erinaceus is guaranteed, and meanwhile the hericium erinaceus is guaranteed to be intact in the grading process.
The conveyor belt is made of food-grade white PU (polyurethane) plastic materials, so that the quality of the hericium erinaceus is guaranteed not to be damaged in the grading process to the maximum extent.
When the type selection of the CCD industrial camera lens, the type selection of the annular light source, the installation condition and the external environment are consistent, the established hericium erinaceus grading model can be applied to hericium erinaceus grading production lines of different scales without modifying the existing grading model, and the online automatic grading of the hericium erinaceus grades is realized.
The invention and its embodiments have been described above schematically, without limitation, and the embodiments shown in the drawings are only one of the embodiments of the invention, and the actual structure and method are not limited thereto, so that those skilled in the art should be informed by the teachings of this invention without departing from the spirit of the invention, and structural embodiments and embodiments similar to this technical solution should fall within the protection scope of the invention.
Claims (5)
1. A hericium erinaceus automatic grading system based on machine vision is characterized by comprising a hardware module, a characteristic parameter extraction and analysis module and a quality grading module which are controlled by an industrial control integrated machine; wherein, the hardware module includes: the device comprises a hericium erinaceus belt type conveying and arranging device, a CCD industrial camera, an annular light source, an image acquisition card, a rotatable circular grading and conveying device, a corresponding grading and conveying channel, an industrial personal computer, an anti-slip baffle, a fixed support and a corresponding connecting and assembling basic part; the characteristic parameter extraction and analysis module collects preset hericium erinaceus appearance characteristic parameters through a hardware module, and the hericium erinaceus appearance characteristic parameters are as follows: the color C of the hericium erinaceus, the shape S of the hericium erinaceus, the integrity I of the hericium erinaceus and the diameter D of the hericium erinaceus; the quality grading module carries out modeling prediction on the acquired hericium erinaceus appearance characteristic parameters through an RGB model, a circularity model, a minimum circumcircle method and a BP neural network model, and then grades of the hericium erinaceus are divided into five grades including a special grade, a first grade, a second grade, a third grade and a grade.
2. A hericium erinaceus grading method is characterized in that: the hericium erinaceus automatic grading system based on machine vision as claimed in claim 1 is adopted for grading, and the grading method is as follows: the hericium erinaceus are arranged before detection through the belt type conveying and arranging device, transmission and detection of single hericium erinaceus can be achieved after arrangement, when the hericium erinaceus pass through the CCD industrial camera and the lower portion of the annular light source through the conveying belt, collection of images of the hericium erinaceus is guaranteed to be completed under the condition that the light source is sufficient, corresponding hericium erinaceus appearance characteristic parameters are obtained, then modeling prediction is conducted through the grading module according to the hericium erinaceus appearance characteristic parameters, and quality grades of the hericium erinaceus are automatically judged; the judged grade information and the real-time position information of the hericium erinaceus are transmitted to the industrial control all-in-one machine, the industrial control all-in-one machine controls the rotary circular grading transmission device to rotate according to the control of the corresponding grade of the hericium erinaceus, the hericium erinaceus are transmitted to the corresponding grade transmission channel, and the whole grading process is completed.
3. The hericium erinaceus grading method according to claim 2, characterized by comprising the following specific steps:
(1) manually grading and recording the quality grade of the hericium erinaceus: the hericium erinaceus is rated by manual observation of a research or a worker who is trained and has rating experience, and the rating standard is as follows: the first level, the second level, the third level, the fourth level and 5 levels outside the levels are respectively recorded by numerical values 1, 2, 3, 4 and 5;
(2) collecting images of the hericium erinaceus: after the hericium erinaceus pass through the belt type conveying and arranging device, the hericium erinaceus sequentially and singly enter an image collecting area, the hericium erinaceus are conveyed to the position under the CCD industrial camera and the annular light source through the conveying belt, the camera collects images of the hericium erinaceus, and the collected images are conveyed to an automatic grading stage of the hericium erinaceus device through a connecting line and an image collecting card;
(3) extracting the quality characteristic parameters of the hericium erinaceus: the hericium erinaceus quality automatic grading software is represented by 4 quality characteristic parameters according to the collected image characteristics;
(4) establishing a hericium erinaceus quality grade database: repeating the steps (1) to (3), and establishing a hericium erinaceus quality grade database, wherein each record of the database consists of 4 characteristic parameters in the step (3) and the hericium erinaceus quality grade in the step (1);
(5) establishing a hericium erinaceus quality grading model: taking each record in a hericium erinaceus quality grade database as a training set, taking 4 characteristic parameters of each record as input, taking hericium erinaceus quality grade as output, and establishing a hericium erinaceus quality grading model according to a modeling method of an RGB model, a circularity model, a minimum circumcircle and a BP neural network model;
(6) automatically grading the quality of the hericium erinaceus: placing hericium erinaceus which are not subjected to manual grading on a conveyor belt of a belt conveyor, collecting image information of the hericium erinaceus to be detected under a CCD industrial camera, transmitting the image information to a hericium erinaceus automatic grading module of the industrial control all-in-one machine, finishing image processing by the hericium erinaceus automatic grading module within 10ms, extracting 4 hericium erinaceus appearance characteristic parameters from the images, and comparing the extracted hericium erinaceus appearance characteristic parameters with the hericium erinaceus quality automatic grading model established in the step (5) to obtain the quality of the hericium erinaceus;
(7) after the hericium erinaceus passes through the visual detection area, the hericium erinaceus enters the classification area, the classification area is provided with a first level, a second level, a third level and five classification conveying channels outside the levels along the conveying belt direction, the executing mechanism adopts a rotatable circular classification conveying device, the industrial control all-in-one machine carries out real-time control on the steering of the conveying belt through the level information output by visual system recognition and the current position information of the hericium erinaceus, the circular rotatable conveying device corresponding to the levels steers, and the hericium erinaceus is conveyed to the classification conveying channels corresponding to the levels.
4. The automatic grading system for hericium erinaceus based on machine vision as claimed in claim 3, wherein: the rating criteria are as follows:
if the grading model is characterized by light yellow, the whole particle is circular without any defect, the diameter is more than 5-6 cm, the grade is 1; if the grading model is characterized by light yellow, the whole particle is circular without any defect, the diameter is 4-5 cm, the grade is two, and the grade is 2; if the grading model is characterized by deep yellow, the whole grain is approximately round without any defect, the diameter is 3-4 cm, the grade is 3; if the grading model is characterized by dark yellow, the whole grain is approximately round and slightly incomplete, the diameter is 1-3 cm, the grade is marked as 4; if the grading model features blackened or whitish, obvious, malformed, incomplete, the grade is outside the grade, and the grade is marked as 5.
5. The method for preparing a hericium erinaceus automatic grading system based on machine vision as claimed in claim 3, wherein: the method for extracting 4 hericium erinaceus appearance characteristic parameters from the image in the step (6) comprises the following steps:
preprocessing the acquired image by combining the image processing software and the grading rule of a grading system with a computer programming language, performing depth processing on the preprocessed image, selecting a proper threshold value, extracting color features of the image through an RGB (red, green and blue) model, and calculating an obtained histogram by using a function calcHist () function carried by Opencv to realize estimation of the color range of the hericium erinaceus; segmenting the background part of the hericium erinaceus image and the hericium erinaceus image by adopting a threshold segmentation method, acquiring whether the shape of the hericium erinaceus tends to be circular or not, and judging the integrity of the hericium erinaceus; and (4) performing edge detection on the contour region of the hericium erinaceus by adopting Canny edge detection, and calculating the diameter of the hericium erinaceus by using a minimum circumcircle method.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210222352.6A CN114378002A (en) | 2022-03-09 | 2022-03-09 | Hericium erinaceus grading system and grading method based on machine vision |
| ZA2023/02019A ZA202302019B (en) | 2022-03-09 | 2023-02-20 | A classification system and method of hericium erinaceus based on machine vision |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210222352.6A CN114378002A (en) | 2022-03-09 | 2022-03-09 | Hericium erinaceus grading system and grading method based on machine vision |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114378002A true CN114378002A (en) | 2022-04-22 |
Family
ID=81205564
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210222352.6A Pending CN114378002A (en) | 2022-03-09 | 2022-03-09 | Hericium erinaceus grading system and grading method based on machine vision |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN114378002A (en) |
| ZA (1) | ZA202302019B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115008512A (en) * | 2022-06-10 | 2022-09-06 | 北京周氏时珍堂药业有限公司 | Automatic slicing and grading equipment for medicinal materials and medicinal material grading method |
| CN115639207A (en) * | 2022-12-26 | 2023-01-24 | 广东省农业科学院设施农业研究所 | Machine vision detection method and system for simultaneously detecting multiple products |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0929185A (en) * | 1995-07-17 | 1997-02-04 | Takaaki Satake | Method for deciding grade of vegetables and fruits having net patterns intrinsic to pericarp surface |
| CN107564000A (en) * | 2017-09-06 | 2018-01-09 | 南京晓庄学院 | Hericium erinaceus Non-Destructive Testing stage division based on computer vision |
| CN110174401A (en) * | 2019-04-30 | 2019-08-27 | 滁州学院 | A kind of the Fresh Grade Breast degree of lignification grading plant and its method of view-based access control model imaging technique |
| CN111114889A (en) * | 2020-01-09 | 2020-05-08 | 广西中烟工业有限责任公司 | Universal sorting device |
| CN211554740U (en) * | 2019-12-26 | 2020-09-22 | 南京帝淮电子科技有限公司 | Universal wheel AGV dolly wireless remote control system |
-
2022
- 2022-03-09 CN CN202210222352.6A patent/CN114378002A/en active Pending
-
2023
- 2023-02-20 ZA ZA2023/02019A patent/ZA202302019B/en unknown
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0929185A (en) * | 1995-07-17 | 1997-02-04 | Takaaki Satake | Method for deciding grade of vegetables and fruits having net patterns intrinsic to pericarp surface |
| CN107564000A (en) * | 2017-09-06 | 2018-01-09 | 南京晓庄学院 | Hericium erinaceus Non-Destructive Testing stage division based on computer vision |
| CN110174401A (en) * | 2019-04-30 | 2019-08-27 | 滁州学院 | A kind of the Fresh Grade Breast degree of lignification grading plant and its method of view-based access control model imaging technique |
| CN211554740U (en) * | 2019-12-26 | 2020-09-22 | 南京帝淮电子科技有限公司 | Universal wheel AGV dolly wireless remote control system |
| CN111114889A (en) * | 2020-01-09 | 2020-05-08 | 广西中烟工业有限责任公司 | Universal sorting device |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115008512A (en) * | 2022-06-10 | 2022-09-06 | 北京周氏时珍堂药业有限公司 | Automatic slicing and grading equipment for medicinal materials and medicinal material grading method |
| CN115008512B (en) * | 2022-06-10 | 2024-02-27 | 北京周氏时珍堂药业有限公司 | Automatic slicing and classifying equipment and classifying method for medicinal materials |
| CN115639207A (en) * | 2022-12-26 | 2023-01-24 | 广东省农业科学院设施农业研究所 | Machine vision detection method and system for simultaneously detecting multiple products |
Also Published As
| Publication number | Publication date |
|---|---|
| ZA202302019B (en) | 2023-05-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107486415B (en) | Thin bamboo strip defect online detection system and detection method based on machine vision | |
| CN114378002A (en) | Hericium erinaceus grading system and grading method based on machine vision | |
| CN109724984B (en) | Defect detection and identification device and method based on deep learning algorithm | |
| CN110779937A (en) | Casting product internal defect intelligent detection device | |
| CN105046700B (en) | Fruit surface defect detection method and system based on gamma correction and color classification | |
| CN107944504B (en) | Board recognition and machine learning method and device for board recognition and electronic equipment | |
| CN108719424A (en) | A kind of aquatic products sorting technique and system based on machine vision | |
| CN107643295A (en) | A kind of method and system of the cloth defect on-line checking based on machine vision | |
| CN111006608A (en) | Transparent plate through hole visual detection system and method | |
| CN109115785A (en) | A kind of casting grinding quality determining method, device and its application method | |
| CN110705655A (en) | Tobacco leaf classification method based on coupling of spectrum and machine vision | |
| CN108287010A (en) | A kind of crab multi objective grading plant and method | |
| CN113145492A (en) | Visual grading method and grading production line for pear appearance quality | |
| CN110394307B (en) | Tobacco leaf automated inspection letter sorting system | |
| CN111709333A (en) | Traceability early warning system and health monitoring method based on abnormal excrement of cage-raised chickens | |
| CN106370667A (en) | Visual detection apparatus and method for quality of corn kernel | |
| CN115082719B (en) | Wood quality grading method | |
| CN106248680A (en) | A kind of engine commutator quality detecting system based on machine vision and detection method | |
| CN107036542A (en) | A kind of ring gear internal-and external diameter appearance detecting method and device | |
| CN109693140A (en) | A kind of intelligent flexible production line and its working method | |
| CN212301356U (en) | Wheel hub welding seam visual detection device | |
| CN106295705A (en) | A kind of many colors material screening number system under movement background | |
| CN107891012B (en) | Pearl size and circularity sorting device based on equivalent algorithm | |
| CN207222383U (en) | Plank sorting system | |
| CN109507204B (en) | Chicken breast lignification grading method and device based on curvature detection |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20220422 |
|
| RJ01 | Rejection of invention patent application after publication |