CN115294507B - Dynamic video data identification method based on fruit appearance - Google Patents
Dynamic video data identification method based on fruit appearance Download PDFInfo
- Publication number
- CN115294507B CN115294507B CN202211233955.2A CN202211233955A CN115294507B CN 115294507 B CN115294507 B CN 115294507B CN 202211233955 A CN202211233955 A CN 202211233955A CN 115294507 B CN115294507 B CN 115294507B
- Authority
- CN
- China
- Prior art keywords
- fruit
- fruits
- line
- neural network
- concave
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 235000013399 edible fruits Nutrition 0.000 title claims abstract description 147
- 238000000034 method Methods 0.000 title claims abstract description 38
- 238000005096 rolling process Methods 0.000 claims abstract description 23
- 238000004519 manufacturing process Methods 0.000 claims abstract description 11
- 238000004891 communication Methods 0.000 claims abstract description 6
- 238000013527 convolutional neural network Methods 0.000 claims description 19
- 238000013528 artificial neural network Methods 0.000 claims description 17
- 230000006378 damage Effects 0.000 claims description 13
- 238000001514 detection method Methods 0.000 claims description 12
- 238000012545 processing Methods 0.000 claims description 7
- 239000003086 colorant Substances 0.000 claims description 4
- 238000013507 mapping Methods 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 2
- 238000005516 engineering process Methods 0.000 claims description 2
- 230000006870 function Effects 0.000 claims description 2
- 230000001788 irregular Effects 0.000 claims description 2
- 238000010586 diagram Methods 0.000 description 8
- 231100000241 scar Toxicity 0.000 description 3
- 238000013473 artificial intelligence Methods 0.000 description 2
- 239000002932 luster Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000037380 skin damage Effects 0.000 description 2
- 208000032544 Cicatrix Diseases 0.000 description 1
- 206010063385 Intellectualisation Diseases 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 210000002615 epidermis Anatomy 0.000 description 1
- 238000010297 mechanical methods and process Methods 0.000 description 1
- 230000037387 scars Effects 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- 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
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/68—Food, e.g. fruit or vegetables
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Multimedia (AREA)
- Computing Systems (AREA)
- General Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Databases & Information Systems (AREA)
- Medical Informatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Data Mining & Analysis (AREA)
- Molecular Biology (AREA)
- Computational Linguistics (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Image Analysis (AREA)
Abstract
A dynamic video data identification method based on fruit appearance relates to the technical field of dynamic video data identification, and specifically comprises the following steps: the method comprises the following steps that firstly, a high-definition camera, a high-speed memory, a GPU, a mainboard and a memory form an identification system, and the identification system is connected to a cloud computer through a wired network or a 5G communication module; secondly, forming a concave rolling production line by rollers, enabling high-definition cameras to be arranged on two sides of the concave rolling production line, collecting videos of fruits rolling on the concave rolling production line, transmitting the videos to a cloud computer, and numbering the fruits appearing in the videos in a one-to-one correspondence mode; a third step of giving the three-dimensional stereo map of the fruit the size of a real fruit; and fourthly, generating a two-dimensional code according to the appearance information of the fruits, pasting the two-dimensional code on the fruits, and classifying the fruits according to grades.
Description
Technical Field
The invention relates to the technical field of dynamic video data identification, in particular to a dynamic video data identification method based on fruit appearance.
Background
Application number CN202111189737.9 discloses a fruit sorting method, device and computer readable storage medium, the fruit sorting method includes: acquiring an image of a fruit to be sorted, and adjusting the image of the fruit to be sorted according to a pre-stored image processing algorithm and an image display parameter to obtain a target image; acquiring a binary image of a target image, determining skin damage information of fruits to be sorted according to the binary image, and sorting the fruits to be sorted for the first time according to the skin damage information; and identifying the shape characteristics of the target image and the color characteristics of the pixel points, and sorting the fruits to be sorted for the second time according to the color characteristics and/or the shape characteristics. The method is used for judging the damage of the epidermis of the fruit.
The appearance of the fruit includes product size, shape, color, surface characteristics, freshness, ripeness, freshness, regularity, presence or absence of blemishes and damage. With the progress of society, consumers have higher and higher requirements on the quality of fruits, and sellers also need to grade the appearance of the fruits, so that the economic value is maximized, and the appearance grading of the fruits is a real demand.
The appearance of the fruit is graded, and the fruit is mechanically and manually processed. Although the mechanical method is good for classifying the size of the fruits, the method is not suitable for selecting the shape, color, surface characteristics, freshness, maturity, freshness, uniformity, whether stains or damages exist; the comprehensive judgment of the size, shape, color, surface characteristics, freshness, uniformity, presence or absence of scars and damages of the fruits cannot be efficiently and accurately carried out.
Disclosure of Invention
In view of the above problems, the present invention aims to provide a dynamic video data identification method based on the appearance of fruits, which dynamically collects fruits rolling on a production line by using a high definition camera, synthesizes a three-dimensional stereo image of the fruits, judges the appearance grade of the fruits, and classifies the fruits according to the appearance grade of the fruits.
The technical solution for realizing the purpose of the invention is as follows:
a dynamic video data identification method based on fruit appearance comprises the following specific steps:
the method comprises the following steps that firstly, a high-definition camera, a high-speed memory, a GPU (graphics processing unit), a mainboard and a memory form an identification system, and the identification system is connected to a cloud computer through a wired network or a 5G communication module;
a concave rolling assembly line is formed by the rollers, high-definition cameras are arranged on two sides of the concave rolling assembly line in a separated mode, videos of the fruits rolling on the concave rolling assembly line are collected, the videos are transmitted to a cloud computer, and the fruits appearing in the videos are numbered in a one-to-one correspondence mode;
thirdly, in a cloud computer, performing frame taking processing on the video, continuously grabbing feature lines of the fruits in continuous frame pictures through a convolutional neural network to form a three-dimensional feature line grid, and mapping the colors of the fruits to the three-dimensional feature line grid one by one to form a three-dimensional stereogram of the fruits; positioning the length of the characteristic line of the fruit through a target detection model YOLO of a deep neural network, and endowing the three-dimensional stereo map of the fruit with the size of a real fruit;
and fourthly, generating a two-dimensional code according to the appearance information of the fruits, pasting the two-dimensional code on the fruits, and classifying the fruits according to grades.
Compared with the prior art, the invention has the beneficial effects that:
(1) The appearance classification of the fruits is more objective and accurate, and the efficiency of the appearance classification of the fruits is improved;
(2) The intellectualization of fruit appearance grading is realized;
(3) And the omnibearing rating of the appearance of the fruits is realized.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the description below are only some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a flow chart of a method for identifying dynamic video data based on fruit shape;
FIG. 2 is a schematic diagram of finding a convex peak line A of a dynamic video data identification method based on fruit shape;
FIG. 3 is a schematic diagram of a peak line A of a positioning convex wave of a dynamic video data identification method based on fruit shape;
FIG. 4 is a diagram of a method for identifying dynamic video data based on fruit shape to find a convex crest line B;
FIG. 5 is a schematic diagram of a positioning convex crest line B of a dynamic video data identification method based on fruit shape;
fig. 6 is a schematic diagram of finding a convex peak line C of a dynamic video data identification method based on fruit shape;
FIG. 7 is a schematic diagram of a peak line C of a fruit profile-based dynamic video data recognition method;
FIG. 8 is a diagram of finding a convex crest line D of a dynamic video data identification method based on fruit shape;
fig. 9 is a schematic diagram of a positioning convex crest line D of a dynamic video data identification method based on fruit shape.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. It should be apparent that the described embodiments are only some of the embodiments of the present invention, and not all of them.
Thus, the following detailed description of the embodiments of the invention is not intended to limit the scope of the invention as claimed, but is merely representative of some embodiments of the invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It should be noted that the embodiments of the present invention and the features and technical solutions thereof may be combined with each other without conflict.
It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures.
The present invention will be described in further detail with reference to examples.
Example (b):
as shown in fig. 1 to 9, the present invention provides a dynamic video data identification method based on fruit shape, the method comprises the following specific steps:
step one, a high-definition camera, a high-speed memory, a GPU, a mainboard and a memory form an identification system, and the identification system is connected to a cloud computer through a wired network or a 5G communication module;
secondly, forming a concave rolling production line by using rollers, enabling high-definition cameras to be arranged on two sides of the concave rolling production line, collecting videos of fruits rolling on the concave rolling production line, transmitting the videos to a cloud computer, and numbering the fruits appearing in the videos in a one-to-one correspondence manner;
thirdly, in a cloud computer, performing frame taking processing on the video, continuously grabbing the characteristic lines of the fruits in continuous frame pictures through a convolutional neural network to form three-dimensional characteristic line grids, and mapping the colors of the fruits to the three-dimensional characteristic line grids one by one to form a three-dimensional stereo image of the fruits; positioning the length of the characteristic line of the fruit through a target detection model YOLO (young only look once) of a deep neural network, and endowing the three-dimensional stereo map of the fruit with the size of a real fruit;
and step four, generating a two-dimensional code according to the appearance information of the fruits, pasting the two-dimensional code on the fruits, and classifying the fruits according to grades.
The first step comprises the steps that a high-definition camera, a high-speed memory, a GPU, a mainboard and a memory form an identification system, and the identification system is connected to a cloud computer through a wired network or a 5G communication module.
Preferably, the first step further comprises: the recognition system is provided with an AI program, a convolutional neural network and a deep neural network, and can independently complete recognition work; the identification system is connected to the cloud computer, the identification system transmits the video to the cloud computer, and the cloud computer transmits the calculation result back to the identification system, so that the identification capability of the identification system is improved; one identification system can be provided with a high-speed memory, a GPU, a mainboard, a memory and a plurality of high-definition cameras, and one identification system can also be provided with a high-speed memory, a GPU, a mainboard, a memory and a high-definition camera, and the high-speed memory, the GPU, the mainboard, the memory and the high-definition cameras are flexibly configured according to actual requirements; whether the recognition system is connected with the cloud computer or not depends on the speed and the quantity of the recognized fruits, when the speed and the quantity of the recognized fruits are few, the recognition system can completely meet the recognition requirement, and when the speed and the quantity of the recognized fruits are large, the recognition system cannot meet the recognition requirement, so that the computing power of the cloud computer is needed; the connection mode between the identification system and the cloud computer is a wired network or a 5G communication module, and the connection mode can be flexibly selected according to specific conditions.
In order to better realize the purpose of the invention, the second step: constitute concave type roll assembly line by the roller bearing, high definition digtal camera is separated the concave type roll assembly line both sides gather fruit and are in rolling video on the concave type roll assembly line transmits the video to cloud computer, numbers the fruit that appears in the video one-to-one ground.
Preferably, the second step further comprises: arranging a high-definition camera on a production line for fruit grade classification, continuously collecting videos of rolling fruits on the production line, and automatically focusing the fruits by adopting an AI (artificial intelligence) technology; the recognition system is capable of independently performing the second step; the fruit sorting machine is characterized in that baffles are arranged on two sides of the concave rolling assembly line, each baffle is provided with a switch, and the switch controls the fruit sorting outlet.
In order to better realize the aim of the invention, the third step is as follows: in a cloud computer, performing frame taking processing on the video, continuously grabbing feature lines of the fruits in continuous frame pictures through a convolutional neural network to form a three-dimensional feature line grid, and mapping the colors of the fruits to the three-dimensional feature line grid one by one to form a three-dimensional stereogram of the fruits; and positioning the length of the characteristic line of the fruit through a target detection model YOLO of a deep neural network, and endowing the three-dimensional stereo map of the fruit with the size of a real fruit.
Preferably, step three further comprises: in the frame picture, capturing a convex peak line and a concave peak line of a fruit through a convolutional neural network, and measuring the lengths of the convex peak line and the concave peak line of the fruit through a target detection model YOLO of a deep neural network; according to the fact that the fruits rotate for a whole circle, the convex crest lines and the concave crest lines of the whole circle of the fruits are grabbed through a convolutional neural network, and the lengths of the convex crest lines and the concave crest lines of the whole circle of the fruits are measured through a target detection model YOLO of a deep neural network; the rolling of the fruits is irregular, the convex crest lines and the concave crest lines of a whole circle of the fruits are crossed with each other, and polygons appear at the crossed positions to form grids; the density of the convex and concave crest lines is proportional to the density of the grid; the higher the density of the grid is, the higher the similarity of the three-dimensional stereo image of the fruit and the fruit is; the characteristic line is formed by convex wave crest lines and concave wave crests, and the characteristic line grids are grids formed by the convex wave crest lines and the concave wave crests. The specific process of forming the three-dimensional perspective view of the fruit is briefly described with reference to fig. 2 to 9, in fig. 2, a represents a convex peak line, a convex peak line a is captured in a frame picture by a convolutional neural network, the length of the convex peak line a is located by a target detection model YOLO of a deep neural network, in fig. 3, the convex peak line a is determined by a capture position characteristic of the convolutional neural network, in fig. 4, a convex peak line B is captured in a frame picture by the convolutional neural network, the length of the convex peak line B is located by the target detection model YOLO of the deep neural network, in fig. 5, a convex peak line B is determined by a capture position characteristic of the convolutional neural network, and the positional relationship between the convex peak line a and the convex peak line B is determined, in fig. 6, a convex crest line C is captured by the convolutional neural network in a frame picture, the length of the convex crest line C is determined by a target detection model YOLO of the deep neural network, in fig. 7, the convex crest line C is determined by a capture position feature of the convolutional neural network and determines the position relationship of the convex crest line a, the convex crest line B, and the convex crest line C, in fig. 8, the convex crest line D is captured by the convolutional neural network in a frame picture, the length of the convex crest line D is determined by the target detection model YOLO of the deep neural network, in fig. 9, the convex crest line D is determined by a capture position feature of the convolutional neural network and determines the position relationship of the convex crest line a, the convex crest line B, the convex crest line C, and the convex crest line D; FIGS. 2 to 9 show the peak lines of four convex lines, which are found to determine the position and size of a circle; by analogy, and combining the characteristic line grids, the three-dimensional stereo view of the fruit is also determined; the method comprises the steps of extracting the characteristics of the shape, color, surface characteristic, freshness degree, maturity degree, freshness degree, regularity, mottle and damage of the fruit to a characteristic line grid through a convolutional neural network, and obtaining the shape, color, surface characteristic, freshness degree, maturity degree, freshness degree, regularity, mottle and damage of the fruit from the step of identifying the appearance characteristic of the fruit to the step of identifying the characteristic value of a three-dimensional stereogram of the fruit through a statistical function.
Preferably, step three further comprises: the three-dimensional perspective view of the fruit is endowed with appearance characteristics of the fruit, and the appearance characteristics of the fruit comprise: fruit size, shape, color and luster, surface characteristic, tender degree, maturity, new degree, regularity, whether have the scar and damage, formulate fruit size, shape, color and luster, surface characteristic, tender degree, maturity, new degree, regularity, have the hierarchical standard of scar and damage, train convolution neural network, convolution neural network is specific according to the hierarchical standard judgement of fruit the fruit grade of the three-dimensional stereogram of fruit, it is specific that the three-dimensional stereogram of fruit is exactly a specific fruit.
In order to better realize the aim of the invention, the step four is as follows: the two-dimensional code is generated according to the appearance information of the fruits and is pasted on the fruits, and the fruits are classified according to grades.
Preferably, the information described in the two-dimensional code includes: fruit size, shape, color, surface characteristics, freshness, maturity, freshness, uniformity, presence or absence of stains and damage; the serial number of the fruit corresponds to the information recorded by the two-dimensional code one by one.
The above embodiments are only used for illustrating the invention and not for limiting the technical solutions described in the invention, and although the present invention has been described in detail in the present specification with reference to the above embodiments, the present invention is not limited to the above embodiments, and therefore, any modification or equivalent replacement of the present invention is made; all such modifications and variations are intended to be included herein within the scope of this disclosure and the appended claims.
Claims (9)
1. A dynamic video data identification method based on fruit appearance is characterized by comprising the following specific steps:
the method comprises the following steps that firstly, a high-definition camera, a high-speed memory, a GPU (graphics processing unit), a mainboard and a memory form an identification system, and the identification system is connected to a cloud computer through a wired network or a 5G communication module;
a concave rolling assembly line is formed by the rollers, high-definition cameras are arranged on two sides of the concave rolling assembly line in a separated mode, videos of the fruits rolling on the concave rolling assembly line are collected, the videos are transmitted to a cloud computer, and the fruits appearing in the videos are numbered in a one-to-one correspondence mode;
thirdly, in a cloud computer, performing frame taking processing on the video, continuously grabbing feature lines of the fruit through a convolutional neural network in continuous frame pictures to form a three-dimensional feature line grid, combining the feature line grid, determining a three-dimensional stereogram of the fruit, mapping colors of the fruit to the three-dimensional feature line grid one by one to form the three-dimensional stereogram of the fruit, and extracting features of the shape, color, surface feature, freshness, regularity, mottle and damage of the fruit to the feature line grid through the convolutional neural network, so that the shape, color, surface feature, freshness, regularity, mottle and damage of the fruit can be obtained by identifying the appearance feature of the fruit and identifying the feature value of the three-dimensional stereogram of the fruit, wherein the feature value is obtained through a statistical function, and the shape, color, surface feature, freshness, regularity, mottle and damage of the fruit can be obtained; positioning the length of the characteristic line of the fruit through a target detection model YOLO of a deep neural network, and endowing the three-dimensional stereogram of the fruit with the size of a real fruit;
the fourth step, according to the appearance information generation two-dimensional code of fruit, paste on fruit, the information that the two-dimensional code recorded includes: the fruits are classified according to grades, wherein the fruits are large in size, shape, color, surface characteristics, freshness, maturity degree, freshness, uniformity, and whether stains and damages exist.
2. A method for fruit-profile-based dynamic video data recognition as claimed in claim 1, wherein: in a first step, the recognition system installs an AI program, a convolutional neural network, and a deep neural network.
3. A method for fruit-profile-based dynamic video data recognition as claimed in claim 1, wherein: in the first step, the recognition system transmits the video to the cloud computer, and the cloud computer transmits the calculation result back to the recognition system.
4. A method for fruit-profile-based dynamic video data recognition as claimed in claim 1, wherein: in the second step, a high-definition camera is arranged on the production line for fruit grade classification, videos of rolling fruits on the production line are continuously collected, and the fruits are automatically focused by adopting an AI technology.
5. A method for fruit-profile-based dynamic video data recognition as claimed in claim 1, wherein: in the third step, in the frame picture, the convex crest line and the concave crest line of the fruit are grabbed through the convolutional neural network, and the lengths of the convex crest line and the concave crest line of the fruit are measured through the target detection model YOLO of the deep neural network.
6. The method according to claim 5, wherein the method comprises the following steps: in the third step, according to the fact that the fruit rotates for a whole circle, the convex crest line and the concave crest line of the whole circle of the fruit are grabbed through the convolution neural network, and the lengths of the convex crest line and the concave crest line of the whole circle of the fruit are measured through a target detection model YOLO of the deep neural network.
7. The method according to claim 6, wherein the method comprises the following steps: in the third step, the rolling of the fruit is irregular, the convex crest lines and the concave crest lines of a whole circle of the fruit are crossed with each other, and polygons appear at the crossed parts to form a grid.
8. The method according to claim 7, wherein the method comprises the following steps: in the third step, the density of the convex and concave crest lines is proportional to the density of the mesh.
9. A method for fruit-profile-based dynamic video data recognition as claimed in claim 8, wherein: in the third step, the higher the density of the grid is, the higher the similarity of the three-dimensional stereo image of the fruit and the fruit is.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211233955.2A CN115294507B (en) | 2022-10-10 | 2022-10-10 | Dynamic video data identification method based on fruit appearance |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211233955.2A CN115294507B (en) | 2022-10-10 | 2022-10-10 | Dynamic video data identification method based on fruit appearance |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115294507A CN115294507A (en) | 2022-11-04 |
| CN115294507B true CN115294507B (en) | 2022-12-16 |
Family
ID=83819429
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211233955.2A Active CN115294507B (en) | 2022-10-10 | 2022-10-10 | Dynamic video data identification method based on fruit appearance |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115294507B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101912847A (en) * | 2010-08-02 | 2010-12-15 | 扬州福尔喜果蔬汁机械有限公司 | Fruit grading system and method based on DSP machine vision |
| CN111709489A (en) * | 2020-06-24 | 2020-09-25 | 广西师范大学 | Citrus identification method based on improved YOLOv4 |
| CN112161618A (en) * | 2020-09-14 | 2021-01-01 | 灵动科技（北京）有限公司 | Storage robot positioning and map construction method, robot and storage medium |
| WO2022090705A1 (en) * | 2020-10-27 | 2022-05-05 | Wootzano Limited | Item packing system, end effetor and method of sorting and/or packing vine fruit |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106982359B (en) * | 2017-04-26 | 2019-11-05 | 深圳先进技术研究院 | Binocular video monitoring method and system and computer readable storage medium |
-
2022
- 2022-10-10 CN CN202211233955.2A patent/CN115294507B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101912847A (en) * | 2010-08-02 | 2010-12-15 | 扬州福尔喜果蔬汁机械有限公司 | Fruit grading system and method based on DSP machine vision |
| CN111709489A (en) * | 2020-06-24 | 2020-09-25 | 广西师范大学 | Citrus identification method based on improved YOLOv4 |
| CN112161618A (en) * | 2020-09-14 | 2021-01-01 | 灵动科技（北京）有限公司 | Storage robot positioning and map construction method, robot and storage medium |
| WO2022090705A1 (en) * | 2020-10-27 | 2022-05-05 | Wootzano Limited | Item packing system, end effetor and method of sorting and/or packing vine fruit |
Non-Patent Citations (3)
| Title |
|---|
| Single Fusion Image from Collections of Fruit Views for Defect Detection and Classification;Antonio Albiol等;《sensors》;20220721;第22卷(第14期);1-14 * |
| 基于改进YOLOv5s和迁移学习的苹果果实病害识别方法;孙丰刚等;《农业工程学报》;20220630;第38卷(第11期);171-179 * |
| 基于物联网技术的海南热带水果动态定价研究;汪汇源等;《江西农业科学》;20210120;第49卷(第2期);223-228 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115294507A (en) | 2022-11-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109961049B (en) | Cigarette brand identification method under complex scene | |
| CN104182765A (en) | Internet image driven automatic selection method of optimal view of three-dimensional model | |
| CN101615245A (en) | Expression recognition method based on AVR and enhancing LBP | |
| CN112365503A (en) | Point cloud data-based transformer substation scene segmentation method | |
| CN109740539B (en) | 3D object identification method based on ultralimit learning machine and fusion convolution network | |
| CN109977899B (en) | Training, reasoning and new variety adding method and system for article identification | |
| Ling et al. | Re-visiting discriminator for blind free-viewpoint image quality assessment | |
| CN114170511A (en) | Pavement crack disease identification method based on Cascade RCNN | |
| CN116363505A (en) | Target picking method based on picking robot vision system | |
| CN103679144B (en) | Method for identifying fruits and vegetables in complex environment based on computer vision | |
| Peng et al. | Litchi detection in the field using an improved YOLOv3 model | |
| Zeeshan et al. | A newly developed ground truth dataset for visual saliency in videos | |
| CN115294507B (en) | Dynamic video data identification method based on fruit appearance | |
| Nebel et al. | Are current monocular computer vision systems for human action recognition suitable for visual surveillance applications? | |
| Suwarningsih et al. | Ide-cabe: chili varieties identification and classification system based leaf | |
| Panda et al. | POST-HARVEST GRADING OF CARICA PAPAYA FRUIT USING IMAGE SEGMENTATION AND SOFT COMPUTING. | |
| CN117079125A (en) | Kiwi fruit pollination flower identification method based on improved YOLOv5 | |
| Yimyam et al. | 3D reconstruction and feature extraction for agricultural produce grading | |
| Lestari et al. | Identifying the Quality System of Cocoa Beans to Increase Productivity Using Backpropagation Neural Network Algorithm: A Case Study at Sumber Rejeki Farmers Group, Patuk Gunung Kidul | |
| Mercol et al. | Automatic classification of oranges using image processing and data mining techniques | |
| Prakasa et al. | Automatic region-of-interest selection for corn seed grading | |
| Dhulipalla et al. | VISIBLE IMAGE PROCESSING FOR FEATURE IDENTIFICATION AND CLASSIFICATION OF AGRICULTURAL PRODUCTS-A CRITICAL REVIEW REPORT | |
| Falih et al. | Optimising Yield Estimation for Grapes: Utilising the Sliding Window Technique for Visual Counting of Bunches and Berries. | |
| Tiator et al. | Using semantic segmentation to assist the creation of interactive VR applications | |
| CN116977274B (en) | Intelligent nut detection method and system suitable for mobile production line |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |