CN117115506A - Corn kernel crushing and identifying method and sorting device based on HSV threshold segmentation - Google Patents

Corn kernel crushing and identifying method and sorting device based on HSV threshold segmentation Download PDF

Info

Publication number
CN117115506A
CN117115506A CN202310840360.1A CN202310840360A CN117115506A CN 117115506 A CN117115506 A CN 117115506A CN 202310840360 A CN202310840360 A CN 202310840360A CN 117115506 A CN117115506 A CN 117115506A
Authority
CN
China
Prior art keywords
image
sorting
corn kernel
module
corn
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
Application number
CN202310840360.1A
Other languages
Chinese (zh)
Inventor
姚艳春
崔春晓
耿端阳
林杰
马闯
王发赢
武继达
黄鹏
李晓珂
李永胜
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shandong University of Technology
Original Assignee
Shandong University of Technology
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Shandong University of Technology filed Critical Shandong University of Technology
Priority to CN202310840360.1A priority Critical patent/CN117115506A/en
Publication of CN117115506A publication Critical patent/CN117115506A/en
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
    • G06V10/765—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects using rules for classification or partitioning the feature space
    • 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
    • B07C5/3425—Sorting according to other particular properties according to optical properties, e.g. colour of granular material, e.g. ore particles, grain
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/0464—Convolutional networks [CNN, ConvNet]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
    • G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Evolutionary Computation (AREA)
  • Multimedia (AREA)
  • Artificial Intelligence (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Computing Systems (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Software Systems (AREA)
  • Databases & Information Systems (AREA)
  • Medical Informatics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • Molecular Biology (AREA)
  • General Engineering & Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Sorting Of Articles (AREA)

Abstract

The invention provides a corn kernel crushing and identifying method and a sorting device based on HSV threshold segmentation. The corn kernel crushing and identifying method provides an image slicing method of a color model based on space color threshold segmentation, removes redundant image background interference, cuts the image into an image with corn kernel outline only, and improves the target outline image occupation ratio. The corn kernel crushing and sorting device comprises a vibration feeding module, an image collecting module, a sorting module, a recycling module, a supporting module and a control module, wherein the image collecting module and the sorting module are integrated at the same station, a rack and a sorting plate are connected into a whole, and the crushed kernels are sorted by adopting gear-rack transmission; the material is transmitted to a single grain feed inlet through the vibration feed module, the photoelectric sensor detects corn grains, the industrial camera is triggered to shoot grain images, and broken grains are identified based on the lightweight convolutional neural network. In the embodiment of the disclosure, the device and the image slice processing method can realize accurate and rapid online identification of broken corn kernels.

Description

Corn kernel crushing and identifying method and sorting device based on HSV threshold segmentation
Technical Field
The invention belongs to the technical field of agricultural material detection, and particularly relates to a corn kernel crushing and identifying method and a sorting device based on HSV threshold segmentation.
Background
In the corn threshing process, the threshing device is in rigid contact with corn kernels, so that the corn kernels are damaged and broken to different degrees, and aflatoxin is easy to induce and influence health. As seed grains, breakage, crushing and the like will greatly affect germination rate, emergence rate and seedling stage growth vigor. At present, the traditional corn seed screening is mainly carried out through manual identification and rejection, the long-time screening is easy to cause error selection, and the screening mechanization and automation degree are low.
Deep learning-based methods have been widely used in the identification of corn kernels. In the aspect of image processing, the established identification model generally directly trains the original image shot by an industrial camera, and in practice, corn kernels do not fully occupy most of images, occupy less images, and have more redundant information, so that the identification accuracy is low. In the aspect of the recognition algorithm, the traditional classical convolutional neural network is mostly adopted, the application to the lightweight convolutional neural network is less, the required model training time is longer, the obtained weight file is larger, the recognition speed is slower, and the method is not suitable for the deployment of small mobile equipment.
The existing corn kernel broken image slicing processing is mainly carried out under the conditions of single background and ideal light, generally, binarization processing is carried out on images, the extracted corn kernel outline images are directly cut, the processing method has high requirements on imaging environment, the problems that image information is lost when the corn kernel images are cut can occur, and the like are difficult to apply to actual sorting operation scenes. Therefore, the grain image can be sliced by the HSV threshold segmentation method so as to realize accurate positioning.
In the aspect of a sorting device, an upper computer of the existing corn seed screening device based on machine vision generally uses a computer or an industrial personal computer to receive sensor signals and identify images, a lower computer controls a mechanical device by utilizing a single chip microcomputer or a PLC, seeds are sorted through communication of the upper computer and the lower computer, and the image identification and sorting are in two stations. Therefore, it is necessary to design an integrated control device to reduce the number of hardware and communication time.
Disclosure of Invention
The first technical problem to be solved by the invention is as follows: how to slice the extracted corn kernel image contour, the problems of more redundant information, small target area occupation ratio, randomness of target positions and the like of the image are reduced.
The second technical problem to be solved by the invention is as follows: how to avoid adopting host computer and lower computer combination mode control maize seed broken and sorting unit, solve that the hardware quantity is many, the device is bulky, the structure is complicated, communication time is long, with high costs etc. to the loaded down with trivial details problem of screening of small seed crop seed.
In order to achieve the technical purpose, the invention adopts the following technical scheme:
a corn kernel crushing and identifying method based on HSV threshold segmentation comprises the steps of color mode conversion, threshold segmentation, contour extraction, vertex coordinate acquisition and cutting.
And collecting RGB color original pictures of the crushed corn kernels, and converting the RGB color original pictures into HSV color pictures.
And (5) obtaining a corn kernel binarization image by using space color threshold segmentation.
And calling an OpenCV library function to extract the corn kernel binarized image contour.
And calculating four vertex coordinates of the minimum circumscribed rectangle of the corn kernel outline according to the extracted binarized image outline.
And calculating the image size of the target area only with the corn kernel image according to the coordinates of the vertices of the rectangle circumscribed by the original image, and cutting.
Preferably, the binary image obtained by the spatial color threshold segmentation is specifically: the method comprises the steps of adjusting the sizes of three components of an image H, S, V in an HSV color model, determining the segmentation effect of a corn kernel image, segmenting the whole outline of the corn kernel, recording an image segmentation threshold value, setting the upper limit value and the lower limit value of the three color components of the HSV color image of the broken corn kernel as the image segmentation threshold value, setting the range of an H threshold value to be 0-36, the range of an S threshold value to be 44-255 and the range of a V threshold value to be 7-255.
Preferably, the mask image is acquired according to the obtained threshold range, the image threshold below the lower limit or above the upper limit is set to 0, and the remaining image threshold is set to 255.
The utility model provides a maize seed grain crushing sorting unit based on HSV threshold value segmentation specifically includes that vibration list grain feeding module, control module, supporting module, image acquisition module, sorting module, recovery module.
The vibration single-particle feeding module consists of a feeding hole, a vibration guide rail, a vibration motor, a conveyor belt baffle, a conveyor belt, a falling slide rail and a funnel guide pipe.
The control module consists of a raspberry group, a voltage converter, a stepping motor driver and a photoelectric sensor.
The support module consists of a steel support, a support bottom plate, a hollow steel column, a stepping motor height adjusting plate and angle irons.
The image acquisition module consists of an industrial camera and a light source.
The sorting module consists of a sorting plate, a sorting brush, a rack, a gear and a stepping motor.
The recovery module consists of a collection box.
Preferably, the front part of the supporting bottom plate is mainly provided with a vibrating single-particle feeding device and a control module, and is supported by a steel bracket, and a raspberry group, a voltage converter and a stepping motor driver are fixed in the bracket through bolts, so that a protection effect is achieved.
Preferably, the rear part of the supporting base plate is mainly a recognition and sorting part, and the recognition and sorting station is supported by four hollow steel columns. The station comprises a square space formed by PVC plates, an industrial camera is placed above, a single-grain feed inlet is arranged in front of the industrial camera, a sorting plate and a rack at the rear of the industrial camera are connected into a whole through nanometer glue, and a sorting brush is fixed on the sorting plate through a brush fixing angle iron to form an integral structure. The lower parts of the left side and the right side are provided with collecting boxes which are fixed by fixing angle irons, and broken seeds and intact seeds are respectively collected.
Preferably, the whole device is divided by a partition plate for fixing the funnel guide pipe, and the partition plate is fixed on the bottom plate by a partition plate fixing angle iron. And finally, installing a photoelectric sensor at a single feed inlet of the station front plate, and detecting whether corn kernels are discharged into the station or not so as to judge whether a subsequent device is started to operate or not.
When the device operates, the power is on, the vibrating motor and the conveyor belt of the vibrating single-particle feeding device are started, and meanwhile, the photoelectric sensor is started for real-time detection.
The material batch gets into the feed inlet, and corn kernel passes through vibrating motor and vibrating guide rail, realizes that the interval single grain advances, and is delivered to the funnel pipe through conveyer belt and whereabouts slide rail.
After slipping through the funnel guide pipe, the photoelectric sensor at the front part of the sorting station is identified and selected, whether corn seeds fall into the station is detected, the device does not operate if no corn seeds fall into the station, if the corn seeds slip into the station from the funnel opening, the photoelectric sensor detects signals, the industrial camera at the top can be started, and the industrial camera photographs the corn seeds in the station.
The obtained image is cut and then passes through a trained identification network, a network identification is carried out to obtain a result, if the result is broken grains, a stepping motor is controlled to drive an integrated rack brush sorting plate to move left through a gear, and corn grains are discharged into a left collecting box; if the corn seeds are intact, the stepping motor is controlled to drive the integrated rack brush sorting plate to move right through the gear, corn seeds are discharged from the right collecting box, sorting of the corn seeds is achieved, automatic reset is achieved after the operation is completed, and the photoelectric sensor waits for the next corn seeds to be discharged, and cyclic identification is achieved.
The invention provides a corn kernel crushing and identifying method and a sorting device based on HSV threshold segmentation, which aim at the situation that background information is more in photographing of an industrial camera and influence on images is larger, and provides an image slicing method based on an HSV color model, wherein a photographed original image is accurately cut into images only containing corn kernels, and irrelevant background interference is removed.
Aiming at the problems of high requirements on processor equipment, large calculated amount, low processing task speed and the like of the traditional convolutional neural network, a lightweight convolutional neural network is selected, and an identification model is simplified on the premise of ensuring the identification accuracy.
Aiming at the defects of more hardware and high cost of the control unit of the existing seed sorting device, the integrated control device is provided, a small raspberry group is adopted to receive sensor signals, and a computer image recognition and mechanical sorting part is integrally controlled, so that the number of hardware and communication time are reduced. Simultaneously, to the comparatively complicated problem of sorting unit mechanical part, provide a simple and easy rapid mechanical separation structure, and mechanical transmission adopts rack and pinion structure, and regard the rack back as one face of station, image acquisition unit and separation unit are in a station simultaneously, realize simple quick separation.
The beneficial effects of the invention are concentrated in the following aspects:
1. in a network identification model part, aiming at the problems of more redundant information, small target area occupation ratio, randomness of target positions and the like of the acquired corn kernel image, the corn kernel image slicing processing method for space color (HSV) threshold segmentation is provided, based on the image preprocessing method, photographed RGB color original pictures can be converted into HSV images, the target areas are obtained after threshold segmentation, image clipping is carried out, image background interference is eliminated, images with the whole space occupied by corn kernels are obtained, and the image identification accuracy is improved.
2. The lightweight convolutional neural network suitable for being deployed at the mobile end is selected, so that the size and the computational complexity of the model are reduced, and higher accuracy is maintained. Compared with a classical convolutional neural network, the selected lightweight network model is half lower in model training time, is quicker in response, and saves raspberry group storage space and computing resources.
3. The device adopts raspberry pie integrated control, does not use a method of combining an upper computer with a lower computer, and reduces cost and communication time. Industrial camera shooting, image network identification and sensor signal receiving, and the control of the stepping motor driver is completed by the raspberry pie alone. The image acquisition unit and the sorting unit of the device are uniformly integrated on the identification sorting station, so that the sorting speed is increased, and the running time is reduced. The mechanical transmission part uses gear and rack transmission, and the sorting brush, the rack and the station rear plate are combined into a whole, so that the rapid corn kernel crushing detection and sorting are realized. The automatic sorting is realized by combining the deep learning method, the screening of broken grains and intact grains is completed, and the problems of easy fatigue, low efficiency, high error rate and the like in manual detection are reduced.
In the present invention, the HSV (Value) is an english abbreviation of Hue, saturation, and brightness, and is described herein.
Drawings
FIG. 1 is a flow chart of a corn kernel image slice;
FIG. 2 is a general schematic diagram of a corn kernel crushing and sorting device;
FIG. 3 is a schematic view of a vibratory single feed apparatus;
FIG. 4 is a schematic diagram of a control module of a corn kernel crushing and sorting device;
FIG. 5A is a schematic view of a support module of a corn kernel crushing and sorting device;
FIG. 5B is a top view of the corn kernel crushing and sorting device support module;
FIG. 6 is a schematic diagram of a corn kernel crushing, identifying and sorting station;
FIG. 7A is a schematic view of a rack brush sorting deck;
FIG. 7B is a front view of a rack brush sorting deck;
FIG. 7C is a top view of a rack brush sorting deck;
FIG. 8 is a flow chart of the operation of the corn kernel crushing device;
in the figure 1, a vibrating single-particle feeding device; 101. a feed inlet; 102. a vibrating guide rail; 103. a vibration motor; 104. a conveyor belt baffle; 105. a conveyor belt; 106. a falling slide rail; 2. a funnel catheter; 3. a control module; 301. raspberry pie; 302. a voltage converter; 303. a stepper motor driver; 4. a support module; 401. a steel support; 402. a support base plate; 403. a hollow steel column; 404. a step motor height adjusting plate; 5. a photoelectric sensor; 6. identifying a sorting station; 601. a photographing port; 602. a station top plate; 603. a station front plate; 604. a single-grain feed inlet; 605. a station bottom plate; 8. a rack brush sorting plate; 801. a sorting plate; 802. fixing angle iron by a hairbrush; 803. sorting brushes; 804. a rack; 10. the division plate is fixed with angle iron; 11. a collection box; 12. the collecting box is fixed with angle iron; 13. a gear; 14. a stepper motor.
Detailed Description
The following detailed description of specific embodiments of the invention will be provided. In order to avoid unnecessary detail, well-known structures or functions will not be described in detail in the following embodiments. Approximating language, as used in the following examples, may be applied to create a quantitative representation that could permissibly vary without resulting in a change in the basic function. Unless defined otherwise, technical and scientific terms used in the following examples have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The corn kernel crushing and sorting device works in two parts, wherein the first part is a part for establishing an identification model through data acquisition, and the second part is a part for operating a mechanical device.
The detection method specifically comprises the following steps:
firstly, 500 pieces of broken grains and 500 pieces of intact grains are collected, and the corn grains occupy a small part of the image, so that the image target is cut through the HSV threshold segmentation algorithm.
The specific cropping steps of fig. 1 are as follows, and the original RGB color image is converted into an HSV color image through an algorithm.
Adjusting the sizes of three components of an image H, S, V in an HSV color model, determining the segmentation effect of a corn kernel image, segmenting the whole outline of the corn kernel, recording an image segmentation threshold value, setting the upper limit value and the lower limit value of the three color components of the HSV color image of the broken corn kernel as the image segmentation threshold value, setting the range of an H threshold value to be 0-36, the range of an S threshold value to be 44-255, the range of a V threshold value to be 7-255, obtaining a mask image, setting the image threshold value lower than or higher than the upper limit to be 0, and setting the rest image threshold values to be 255.
And calling an OpenCV library function to extract the outline of the binarized image and obtaining four vertex coordinates of a rectangle with the minimum circumscribed outline of the corn kernel.
And calculating the image size of the target area on the original image according to the vertex coordinates of the circumscribed rectangle, and cutting to obtain a final slice image. And calculating the cut image through a lightweight convolutional neural network, obtaining an identification model capable of identifying corn kernels through migration learning, and disposing the identification model in a raspberry group to serve as a visual identification part for device operation.
How the present invention can be implemented will be further described below with reference to the structure of the detecting device.
As shown in fig. 2, the vibration motor 103 and the conveyor belt 105 of the vibration single-particle feeding device 1 are started after the power-on start, and the photoelectric sensor 5 is started for real-time detection.
As shown in fig. 3, the materials enter the feeding port 101 in batches, corn kernels are subjected to single grain advance at intervals by the vibrating motor 103 and the vibrating guide rail 102, and are conveyed to the funnel guide pipe 2 by the conveyor belt 105 and the falling slide rail 106.
As shown in fig. 2 and 5A, after the corn kernels slide down through the funnel guide pipe 2, the photoelectric sensor 5 at the single kernel feeding hole 604 at the front part of the recognition and sorting station 6 detects whether corn kernels fall down, if no kernels fall down, the device does not operate, if corn kernels slide into the station from the funnel hole, the photoelectric sensor detects a signal, the industrial camera 7 at the top is started, and the industrial camera photographs the corn kernels in the station.
As shown in fig. 1 and fig. 4, the obtained image is cut, sent to an identification network deployed in a raspberry group 301, and identified by the network to obtain a result, if the image is broken, the stepping motor 14 is controlled to drive the integrated rack brush sorting plate 8 to move left through the gear 13 to discharge corn kernels into the left collecting box 11, if the image is perfect, the stepping motor 14 is controlled to drive the integrated rack brush sorting plate 8 to move right through the gear 13 to discharge the corn kernels into the right collecting box 11, so that sorting of the corn kernels is realized, automatic reset is performed after the action is completed, and the photoelectric sensor waits for the discharge of the next corn kernels to be circularly identified.
The present invention is not limited to the above-mentioned embodiments, and any changes or substitutions that can be easily understood by those skilled in the art within the technical scope of the present invention are intended to be included in the scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

Claims (11)

1. A corn kernel crushing and identifying method based on HSV threshold segmentation is characterized in that:
the method comprises the steps of color mode conversion, threshold segmentation, contour extraction, vertex coordinate acquisition and clipping;
collecting RGB color original pictures of broken corn kernels, and converting the RGB color original pictures into HSV color pictures;
dividing by using a spatial color threshold to obtain a corn kernel binarization image;
calling an OpenCV library function to extract a corn kernel binarization image contour;
calculating four vertex coordinates of the minimum circumscribed rectangle of the corn kernel outline according to the extracted binarized image outline;
and calculating the image size of the target area only with the corn kernel image according to the coordinates of the vertices of the rectangle circumscribed by the original image, and cutting.
2. The threshold segmentation in the corn kernel breaking recognition method based on HSV threshold segmentation according to claim 1, wherein three components of an image H, S, V in an HSV color model are adjusted, the segmentation effect of a corn kernel image is determined, the whole outline of the corn kernel is segmented, the image segmentation threshold is recorded, the upper limit value and the lower limit value of three color components of the HSV color image of the broken corn kernel are set as the threshold of image segmentation, the range of H threshold is set to be 0-36, the range of S threshold is set to be 44-255, the range of V threshold is set to be 7-255, a mask image is acquired, the image threshold lower than or higher than the upper limit is set to be 0, and the rest image threshold is set to be 255.
3. Corn kernel crushing and sorting device based on HSV threshold segmentation specifically comprises:
the device comprises a vibration single-grain feeding module, a control module, a supporting module, an image acquisition module, a sorting module and a recycling module;
the vibration feeding module consists of a feeding hole (101), a vibration guide rail (102), a vibration motor (103), a conveyor belt baffle (104), a conveyor belt (105), a falling slide rail (106) and a funnel guide pipe (2);
the control module consists of a raspberry group (301), a voltage converter (302), a stepping motor driver (303) and a photoelectric sensor (5);
the support module consists of a steel support (401), a support bottom plate (402), a hollow steel column (403), a stepping motor height adjusting plate (404) and angle irons;
the image acquisition module consists of an industrial camera (7) and a light source;
the sorting module consists of a sorting plate (801), a sorting brush (803), a rack (804), a gear (13) and a stepping motor (14);
the recovery module consists of a collection box (11).
4. A corn kernel crushing and sorting device based on HSV threshold segmentation according to claim 3, characterized in that the front part of the supporting base plate (402) is mainly provided with a vibrating single-grain feeding device (1) and a control module (3), and is supported by a steel bracket (401), and a raspberry pie (301), a voltage converter (302) and a stepping motor driver (303) are fixed inside the bracket through bolts to play a role of protection;
the rear part of the corn kernel sorting device is provided with an image acquisition module and a sorting module, the middle of the corn kernel crushing and sorting device is separated by a separation plate (9), and the separation plate is fixed on a bottom plate by a separation plate fixing angle iron (10) and is used for fixing a funnel guide pipe (2).
5. The corn kernel crushing and sorting device based on HSV threshold segmentation according to claim 3, wherein a photoelectric sensor (5) is fixedly arranged at a single-kernel feed inlet (604) of a station front plate (603) through bolts, the photoelectric sensor (5) is in a diffuse reflection mode, faces to a funnel duct outlet, a corn kernel entering station signal is detected, and an industrial camera is triggered to shoot a kernel image.
6. The corn kernel crushing and sorting device based on HSV threshold segmentation according to claim 3, specifically comprising:
vibration single-grain feeding device (1), funnel guide pipe (2), control module (3), support module (4), photoelectric sensor (5), discernment select separately station (6), industry camera (7), rack brush sorting board (8), division board (9), division board fixed angle bar (10), collecting box (11), collecting box fixed angle bar (12), gear (13) and step motor (14).
7. The vibrating single grain feeding device (1) in the corn grain crushing and sorting device based on HSV threshold segmentation according to claim 6, specifically comprising:
the device comprises a feed inlet (101), a vibrating guide rail (102), a vibrating motor (103), a conveyor baffle (104), a conveyor (105) and a falling slide rail (106);
the vibration single-grain feeding device (1) is fixed on a steel support (401) through a bolt link, a vibration guide rail (102) is made of PVC plates, the vibration guide rail is fixed on a vibration motor (103) through a bolt, and a conveyor belt baffle (104) is fixed on a conveyor belt (105) through angle irons.
8. The control module (3) in a corn kernel crushing and sorting device based on HSV threshold segmentation according to claim 6, characterized in that it comprises in particular:
the raspberry pie (301), the voltage converter (302) and the stepping motor driver (303) are all fixed on the front part of the supporting bottom plate (402) through bolts.
9. The support module (4) in a corn kernel crushing and sorting device based on HSV threshold segmentation according to claim 6, characterized in that it comprises in particular:
the steel support (401), supporting baseplate (402), hollow steel column (403), step motor height adjusting plate (404), hollow steel column (403) are fixed on supporting baseplate (402) through the bolt link, step motor height adjusting plate (404) are fixed on supporting baseplate (402) through the steel nail.
10. The identification and sorting station (6) in a corn kernel crushing and sorting device based on HSV threshold segmentation according to claim 6, characterized in that the material is PVC plate and is connected by angle iron, comprising in particular:
photographing port (601), station top plate (602), station front plate (603), single feed port (604) and station bottom plate (605).
11. The rack brush sorting deck (8) in a corn kernel crushing and sorting apparatus based on HSV threshold segmentation of claim 6, characterized by specifically comprising:
sorting board (801), fixed angle bar of brush (802), sorting brush (803), rack (804) and sorting board (801) link as an organic wholely through nano glue, and control sorting brush is fixed on sorting board (801) through the angle bar.
CN202310840360.1A 2023-07-11 2023-07-11 Corn kernel crushing and identifying method and sorting device based on HSV threshold segmentation Pending CN117115506A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310840360.1A CN117115506A (en) 2023-07-11 2023-07-11 Corn kernel crushing and identifying method and sorting device based on HSV threshold segmentation

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310840360.1A CN117115506A (en) 2023-07-11 2023-07-11 Corn kernel crushing and identifying method and sorting device based on HSV threshold segmentation

Publications (1)

Publication Number Publication Date
CN117115506A true CN117115506A (en) 2023-11-24

Family

ID=88793619

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310840360.1A Pending CN117115506A (en) 2023-07-11 2023-07-11 Corn kernel crushing and identifying method and sorting device based on HSV threshold segmentation

Country Status (1)

Country Link
CN (1) CN117115506A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117876466A (en) * 2024-03-13 2024-04-12 浙江托普云农科技股份有限公司 Corn ear phenotype parameter calculation method, system and device based on vision

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117876466A (en) * 2024-03-13 2024-04-12 浙江托普云农科技股份有限公司 Corn ear phenotype parameter calculation method, system and device based on vision

Similar Documents

Publication Publication Date Title
CN106733701B (en) Potato multi-stage combination dysnusia detecting system and detection method
CN101905215B (en) Digitalized paddy rice seed testing machine
CN202021164U (en) Peanut exterior quality detecting and sorting device
CN113820325B (en) Corn grain direct-harvest impurity-containing rate and breakage rate online detection system and method
CN117115506A (en) Corn kernel crushing and identifying method and sorting device based on HSV threshold segmentation
CN105806751A (en) On-line monitoring system and method for crushing of cereals in grain tank of combine harvester
CN111359907B (en) Real-time online detection system and method for wheat scab infection particle rate based on embedded deep learning
CN109085170A (en) Synchronous belt type corn kernel percentage of damage on-line measuring device and method
CN103706574A (en) Automatic solid grain sorting system
CN109030504A (en) The panorama potato defect detecting device and method combined based on actual situation imaging
CN110584164A (en) Green plum sorting production line and sorting method
CN207222383U (en) Plank sorting system
CN110575973B (en) Crop seed quality detection and screening system
Song et al. Design and experiment of a sorting system for haploid maize kernel
CN207147533U (en) A kind of Miniature digital rice species test machine
CN114839197B (en) Rice damage detection device and detection method
CN215613293U (en) Seed sorting device based on machine vision
CN114140675B (en) Sugarcane seed screening system and method based on deep learning
CN214794500U (en) Online detection device for wheat appearance quality
CN115861784A (en) Fish feed feeding method and fish feed feeding amount analysis system
CN108993927B (en) Crop seed grain separating equipment and control method thereof
Gan et al. Design and testing of a machine-vision-based air-blow sorting platform for famous tea fresh leaves production
CN113791078A (en) Method and device for batch detection of internal cracks of corn seeds
CN114993942B (en) Full-automatic rice seed testing equipment and method
CN114354628B (en) Rhizome agricultural product defect detection method based on machine vision

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