CN116778474A - Intelligent phenotype analyzer for tomato fruits - Google Patents
Intelligent phenotype analyzer for tomato fruits Download PDFInfo
- Publication number
- CN116778474A CN116778474A CN202310649762.3A CN202310649762A CN116778474A CN 116778474 A CN116778474 A CN 116778474A CN 202310649762 A CN202310649762 A CN 202310649762A CN 116778474 A CN116778474 A CN 116778474A
- Authority
- CN
- China
- Prior art keywords
- fruit
- intelligent
- tomato
- module
- camera
- 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
- 235000013399 edible fruits Nutrition 0.000 title claims abstract description 95
- 235000007688 Lycopersicon esculentum Nutrition 0.000 title claims abstract description 61
- 240000003768 Solanum lycopersicum Species 0.000 title claims description 57
- 230000011218 segmentation Effects 0.000 claims abstract description 30
- 230000001360 synchronised effect Effects 0.000 claims abstract description 30
- 238000012545 processing Methods 0.000 claims abstract description 27
- 238000013135 deep learning Methods 0.000 claims abstract description 18
- 238000012937 correction Methods 0.000 claims abstract description 7
- 230000033001 locomotion Effects 0.000 claims description 13
- 238000004458 analytical method Methods 0.000 claims description 7
- 238000011176 pooling Methods 0.000 claims description 7
- 101100295091 Arabidopsis thaliana NUDT14 gene Proteins 0.000 claims description 4
- 230000005540 biological transmission Effects 0.000 claims description 3
- 238000003702 image correction Methods 0.000 claims description 3
- 238000005516 engineering process Methods 0.000 abstract description 9
- 210000001113 umbilicus Anatomy 0.000 abstract description 7
- 241000227653 Lycopersicon Species 0.000 abstract 4
- 238000009395 breeding Methods 0.000 description 11
- 230000001488 breeding effect Effects 0.000 description 11
- 238000003709 image segmentation Methods 0.000 description 9
- 238000000034 method Methods 0.000 description 9
- 238000011161 development Methods 0.000 description 7
- 230000018109 developmental process Effects 0.000 description 7
- 238000001514 detection method Methods 0.000 description 6
- 239000003086 colorant Substances 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000012549 training Methods 0.000 description 3
- 238000013473 artificial intelligence Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000012952 Resampling Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000000105 evaporative light scattering detection Methods 0.000 description 1
- 235000013305 food Nutrition 0.000 description 1
- 235000012041 food component Nutrition 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000877 morphologic effect Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Landscapes
- Image Analysis (AREA)
Abstract
The invention provides an automatic, intelligent and high-precision intelligent tomato fruit phenotype analyzer based on deep learning and image processing technology, which comprises a data acquisition module, a control module and a data processing module, wherein the data acquisition module comprises a camera and a triaxial synchronous belt precise sliding table module, the camera is arranged on a sliding block of the triaxial synchronous belt precise sliding table module, and the data processing module comprises a fruit precise identification system, a fruit different structure segmentation system and an image distortion correction system. The invention provides an automatic, intelligent and high-precision intelligent tomato fruit phenotype analyzer based on deep learning and image processing technology, which utilizes a fruit precise identification system developed based on a PaddlePaddle deep learning platform to precisely identify pedicel, umbilicus and the like of tomato fruits with different forms, and adopts a fruit different-structure segmentation system based on a deep Labv3 algorithm to precisely segment different fruit structures of tomatoes.
Description
Technical Field
The invention relates to the technical field of agricultural breeding, in particular to an intelligent phenotype analyzer for tomato fruits.
Background
At present, intelligent agriculture is an important embodiment of modern agriculture development in China, and is benefited from the vigorous development of genomics, phenotypic group science and artificial intelligence, and the core strategic appeal of the intelligent agriculture by the nation, intelligent breeding is taken as an important branch of the intelligent agriculture, and the intelligent breeding becomes a hot word in recent years. Breeding is a process of selecting genotype materials with optimal various phenotypic indicators under given environmental conditions. Intelligent breeding is a process of accelerating breeding material screening by utilizing artificial intelligence technology to help breeders.
In the traditional breeding work, the measurement work of the diameter, the area, the size, the color and the components of the internal cavity and the like of the circumscribed circle of the fruit are mainly finished manually. The method has the advantages of large labor force, high labor cost and low efficiency, and is not suitable for mass detection. At present, china has entered an intelligent breeding age, and a plurality of intelligent phenotypic analysis devices aiming at mainstream food crops are available on the market, and few intelligent phenotypic analysis devices aiming at tomato fruits are available. Tomato breeding quality detection in addition to detection of nutritional components and the like, fruit phenotype is also an important parameter for measuring breeding quality. The accuracy of the phenotype data depends on the segmentation of different structures of tomato fruits, but the colors of different varieties of tomatoes are quite different, the colors of different structures of the fruits are similar, and the traditional image processing method is difficult to segment the different structures.
Disclosure of Invention
Aiming at the problem of lack of intelligent phenotype analysis equipment of the existing tomato fruits, the invention provides an automatic, intelligent and high-precision intelligent phenotype analyzer for tomato fruits based on deep learning and image processing technology.
The invention solves the technical problems by adopting the scheme that: the intelligent tomato fruit phenotype analyzer comprises a data acquisition module, a control module and a data processing module, wherein the data acquisition module comprises a camera and a triaxial synchronous belt precise sliding table module, and the camera is arranged on a sliding block of the triaxial synchronous belt precise sliding table module.
The controller module comprises a motion control system, the motion control system is connected with and controls the three-axis synchronous belt precise sliding table module, the three-axis synchronous belt precise sliding table module drives the camera to photograph tomatoes, and the camera uploads image data to the data processing module for analysis and processing.
The data processing module comprises a fruit accurate identification system, a fruit different structure segmentation system and an image distortion correction system.
The image uploaded by the camera firstly enters an image distortion correction system to carry out image correction, corrected image data are respectively transmitted to a fruit accurate identification system and a fruit different structure segmentation system to carry out tomato feature identification, and feature data are output by the tomato fruit accurate identification system and the fruit different structure segmentation system to obtain phenotype features of tomatoes.
The fruit accurate identification system is based on a PaddlePaddle deep learning platform and is used for accurately identifying pedicel, peduncles, fruits, fruit walls and umbilicus of tomato fruits.
The fruit different structure segmentation system constructs a fruit different structure segmentation algorithm based on the deep Labv3 algorithm, the deep Labv3 algorithm modifies the ASPP module, global average pooling capturing global information is increased, and different fruit structures of tomatoes are accurately segmented.
The three-axis synchronous belt precise sliding table module comprises a group of parallel sliding rails, the inner ends of the group of sliding rails are connected through a synchronous connecting rod in a transmission way, transverse rails are arranged on the parallel sliding rails, sliding blocks are arranged on the transverse rails, a camera is arranged on the sliding blocks, and the three-axis synchronous belt precise sliding table module is driven by a servo motor.
The invention has the beneficial effects that: the invention provides an automatic, intelligent and high-precision intelligent tomato fruit phenotype analyzer based on deep learning and image processing technology, which utilizes a fruit precise identification system developed based on a PaddlePaddle deep learning platform to precisely identify pedicel, umbilicus and the like of tomato fruits with different forms, and adopts a fruit different-structure segmentation system based on a deep Labv3 algorithm to precisely segment different fruit structures of tomatoes.
The image distortion correction program of the intelligent detector is used for correcting the image of the oblique projection error existing between the camera lens and the fruit to be detected, so that the accuracy and resolution of image acquisition are improved, and the later image processing work is facilitated.
The three-axis synchronous belt precise sliding table module is combined with the motion control system, so that the camera can collect the image information of the object to be collected at a certain designated position with high efficiency and high precision.
Aiming at the problem of detecting morphological structure phenotype characters of tomato fruits, the invention develops an automatic, intelligent and high-precision intelligent phenotype analyzer for tomato fruits by applying the advanced deep learning and image processing technology, and can finish the operation more efficiently.
Drawings
Fig. 1 is a schematic perspective view of the present invention.
Fig. 2 is a block diagram of the present invention.
Fig. 3 is a schematic structural diagram of the three-axis synchronous belt precise sliding table module.
Fig. 4 is an image segmentation flow diagram of an image segmentation kit.
Fig. 5 is a flow chart of deep lab v3 algorithm image segmentation.
Reference numerals in the drawings: the device comprises a shell 1, a data acquisition area 11, a control area 12, a triaxial synchronous belt precise sliding table module 10, a sliding rail 101, a transverse track 102, a camera 103, a sliding block 104 and a synchronous connecting rod 105.
Description of the embodiments
The invention will be further described with reference to the drawings and examples.
Example 1: in order to fill the lack of intelligent phenotype analysis equipment for tomato fruits, the invention provides an automatic, intelligent and high-precision intelligent phenotype analyzer for tomato fruits based on deep learning and image processing technology.
The intelligent tomato fruit phenotype analyzer comprises a data acquisition module, a control module and a data processing module, wherein the data acquisition module comprises a camera and a triaxial synchronous belt precise sliding table module, and the camera is arranged on a sliding block of the triaxial synchronous belt precise sliding table module.
The controller module comprises a motion control system, the motion control system is connected with and controls the three-axis synchronous belt precise sliding table module, the three-axis synchronous belt precise sliding table module drives the camera to photograph tomatoes, and the camera uploads image data to the data processing module for analysis and processing.
Based on the prior art, a set of motion control system is provided, so that the camera can collect the image information of an object to be collected at a certain designated position with high efficiency and high precision. The high-speed and high-precision servo motor is adopted to provide power for the triaxial synchronous belt precise sliding table module, and the camera can be matched with a high-efficiency motion control system to efficiently and precisely acquire the image information of the object to be acquired, and the motion position error is in the range of +/-0.05 mm overall.
The data processing module comprises a fruit accurate identification system, a fruit different structure segmentation system and an image distortion correction system.
The image uploaded by the camera firstly enters an image distortion correction system to carry out image correction, corrected image data are respectively transmitted to a fruit accurate identification system and a fruit different structure segmentation system to carry out tomato feature identification, and feature data are output by the tomato fruit accurate identification system and the fruit different structure segmentation system to obtain phenotype features of tomatoes.
Because the placement positions of all the collecting objects are different, the camera lens is not perpendicular to each collecting object during collecting, and therefore oblique projection errors exist between the camera lens and the tomato fruit body, and the measurement results are inaccurate due to the errors. In addition, the skin colors of different acquisition objects are different, reflection and other phenomena can sometimes occur, and the invention provides an image distortion correction program suitable for an intelligent detector on the basis of words.
The fruit accurate identification system is based on a PaddlePaddle deep learning platform and is used for accurately identifying pedicel, peduncles, fruits, fruit walls and umbilicus of tomato fruits.
The deep learning has the advantages of high accuracy, strong adaptability and the like. The PaddlePaddle deep learning platform has the advantages of simple and stable design, high speed and the like. Aiming at the recognition problems of pedicel, peduncles, umbilicus and the like of the tomato fruits, the most advanced deep learning technology in China is adopted at present, a precise recognition algorithm for the pedicel, peduncles, fruits, walls and umbilicus of the tomato fruits based on deep learning is developed, and the tomato fruits are precisely recognized.
The fruit different structure segmentation system constructs a fruit different structure segmentation algorithm based on the deep Labv3 algorithm, and the fruit has different structures and has a plurality of problems when the fruit is segmented in structure. ASPP in the deep Labv3 algorithm can be used for solving the problem of size difference of different detection targets, and resampling can be effectively carried out by using hole convolution of different conditions on a given feature layer, and convolution kernels of different receptive fields are constructed to obtain multi-scale object information. Based on the problems, the invention adopts the deep Labv3 algorithm widely applied at home and abroad and in industry to construct different fruit structure segmentation algorithms, the deep Labv3 algorithm modifies the ASPP module, global average pooling is added so as to better capture global information, and different fruit structures of tomatoes can be accurately segmented.
In view of the fact that the traditional image processing algorithm, such as openCV, has a poor effect on similar color processing, is difficult to be used for image segmentation of different structures (pedicel, fruit wall and the like) of tomatoes, the deep Lab v3 algorithm of the hundred-degree company paddledle deep learning platform is adopted for segmentation of different structures of tomatoes.
As shown in fig. 4 and fig. 5, paddleSeg is an end-to-end image segmentation kit based on a flying oar PaddlePaddle, a 45+ model algorithm and a 140+ pre-training model are built in, a configuration driving and API call development mode is supported, the whole flow of data marking, model development, training, compression and deployment is opened, four segmentation capacities of semantic segmentation, interactive segmentation, marking and panoramic segmentation are provided, and a boosting algorithm is applied to the ground in the scenes of medical treatment, industry, remote sensing, entertainment and the like.
Deep lab v3 is a deep learning model used for image segmentation tasks. The deep Lab v3 structure adopts the technologies of cavity volume, multi-scale pyramid pooling and the like, so that the accuracy of image segmentation can be improved while the high resolution is maintained. The core of the deep lab v3 structure is the hole convolution. The traditional convolution operation only considers the relation among local pixels, and the cavity convolution can enlarge the receptive field of the convolution kernel under the condition of not increasing the number of parameters. Thus, the context information in the image can be better captured, and the segmentation accuracy is improved. Another important technique is multi-scale pyramid pooling. This technique can pool images at different scales to capture features at different scales. Thus, the size and shape change of the object in the image can be better processed, and the segmentation robustness is improved. In addition to hole convolution and multi-scale pyramid pooling, the deep lab v3 architecture also employs other techniques such as spatial pyramid pooling, deconvolution, and conditional random fields. These techniques may further improve the accuracy and robustness of the segmentation.
Specific principle of image segmentation: firstly, a large number of tomato images are collected, simple data cleaning (namely, images which do not meet the training requirement are removed), the collected images are marked by using a lableme tool, the marked images are trained into different models through a paldlefilled deep learning platform, and after the accuracy of the models meets the requirement, the different structures of any tomatoes can be accurately segmented through the trained models. The method has high stability, and can eliminate the influence of light variation and tomato color.
The three-axis synchronous belt precise sliding table module comprises a group of parallel sliding rails, wherein the inner ends of the group of sliding rails are in transmission connection through synchronous connecting rods, transverse rails are arranged on the parallel sliding rails, sliding blocks are arranged on the transverse rails, longitudinal shaft rails are arranged on the sliding blocks, cameras are arranged on the longitudinal shaft rails, and the three-axis synchronous belt precise sliding table module is driven by a servo motor.
The three-axis synchronous belt precise sliding table module consists of three linear motion axes which are an X axis, a Y axis and a Z axis respectively, wherein a group of parallel sliding rails are used as the Y axis to support the camera to move back and forth, the transverse rail is used as the X axis to drive the camera to move left and right, and finally the longitudinal axis rail arranged on the transverse rail is used as the Z axis to control the camera to move up and down, so that different distance visual angles are provided for the analyzer.
The intelligent tomato fruit phenotype analyzer provided by the invention comprises a shell 1, wherein the shell is divided into two areas, a data acquisition area 11 and a control area 12, a triaxial synchronous belt precise sliding table module 10 is arranged in the data acquisition area 11 of the shell 1, a tray is arranged at the bottom of the data acquisition area 11 and is used for containing tomatoes to be detected, a controller and data processing equipment are arranged in the control area 12, the controller and the data processing equipment adopt an Injetson NANO development board, and the Injetson NANO development board is based on an ARM Cortex-A7 quad processor, a Mali-G52 GPU and a 1GB LPDDR4X memory development board and can be used for developing a real-time operating system and Internet of things application.
The intelligent tomato fruit phenotype analyzer also comprises a display.
The invention has simple operation, an operator firstly places the acquisition object in the instrument before using, presses the switch key and carries out related operation through the touch display screen on the right side of the equipment. And the collected phenotype data is analyzed and counted by using matched analysis software, so that later breeding work is facilitated.
In hardware, a high-definition industrial camera and a high-precision servo motor are adopted, and the high-definition industrial camera and the high-precision servo motor are combined, so that the camera can accurately collect images at a certain designated position. On the software, a tomato fruit pedicel, fruit peduncles, fruits, fruit walls and fruit umbilicus accurate identification detection algorithm based on deep learning is developed; adopting a deep Labv3 algorithm to construct different structure segmentation algorithms of fruits; developing an image distortion correction program suitable for the intelligent detector; a set of motion control systems has also been developed.
The invention adopts the acquisition scheme of single-camera spatial movement to acquire crop images, develops a camera distortion correction program, can acquire the characters of various tomato fruits, improves the accuracy and resolution of image acquisition, and is convenient for later image processing work.
The invention not only can complete a series of tasks of identification, image segmentation, data collection and processing and the like of various tomato fruit collection objects, but also solves the problems of traditional fruit sorting and detection and the like, and can complete the full-process unmanned detection of tomato fruit phenotype parameters more automatically, intelligently and with high precision.
The invention has the advantages of fully automatic operation of the whole machine, great reduction of labor force, saving of manpower resource cost and improvement of working efficiency. Meanwhile, the program of the device is modified, so that the characteristics of other small fruits can be collected, and the application range is wider.
The foregoing has outlined the basic principles, features, and advantages of the present invention. It will be understood by those skilled in the art that the present invention is not limited to the embodiments described above, and that the above embodiments and descriptions are merely illustrative of the principles of the present invention, and various changes and modifications may be made therein without departing from the spirit and scope of the invention, which is defined by the appended claims.
Claims (5)
1. An intelligent phenotype analyzer for tomato fruits is characterized by comprising a data acquisition module, a control module and a data processing module,
the data acquisition module comprises a camera and a triaxial synchronous belt precise sliding table module, and the camera is arranged on a sliding block of the triaxial synchronous belt precise sliding table module;
the controller module comprises a motion control system, the motion control system is connected with and controls the three-axis synchronous belt precise slipway module, the three-axis synchronous belt precise slipway module drives the camera to photograph tomatoes, and the camera uploads image data to the data processing module for analysis and processing;
the data processing module comprises a fruit accurate identification system, a fruit different structure segmentation system and an image distortion correction system;
the image uploaded by the camera firstly enters an image distortion correction system to carry out image correction, corrected image data are respectively transmitted to a fruit accurate identification system and a fruit different structure segmentation system to carry out tomato feature identification, and feature data are output by the tomato fruit accurate identification system and the fruit different structure segmentation system to obtain phenotype features of tomatoes.
2. The intelligent tomato fruit phenotyping analyzer of claim 1, wherein the precise fruit identification system is based on a paddlepad deep learning platform for precise identification of tomato fruit pedicel, fruit wall, fruit navel.
3. The intelligent phenotype analyzer for tomato fruits according to claim 1, wherein the different structure segmentation system constructs different structure segmentation algorithms for the fruits based on deep labv3 algorithm, the deep labv3 algorithm modifies ASPP modules, global average pooling is added to better capture global information, and accurate segmentation is performed for different fruit structures of the tomatoes.
4. Intelligent tomato fruit phenotype analyzer according to claim 1, characterized in that the triaxial synchronous belt precision slipway module comprises a group of parallel slide rails (101), the inner ends of the group of slide rails (101) are connected in a transmission way through a synchronous connecting rod (105), a transverse rail (102) is arranged on the parallel slide rails, a sliding block (104) is arranged on the transverse rail, and a camera (103) is arranged on the sliding block (104).
5. The intelligent tomato fruit phenotype analyzer of claim 4, wherein the three-axis synchronous belt precise slipway module is driven by a servo motor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310649762.3A CN116778474A (en) | 2023-06-02 | 2023-06-02 | Intelligent phenotype analyzer for tomato fruits |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310649762.3A CN116778474A (en) | 2023-06-02 | 2023-06-02 | Intelligent phenotype analyzer for tomato fruits |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116778474A true CN116778474A (en) | 2023-09-19 |
Family
ID=87992226
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310649762.3A Pending CN116778474A (en) | 2023-06-02 | 2023-06-02 | Intelligent phenotype analyzer for tomato fruits |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116778474A (en) |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE202005014072U1 (en) * | 2005-09-06 | 2005-11-24 | Maier, Florian | Apparatus photographing fixed objects from various perspectives, includes camera mounting plate on carriage and rails with computer control over relative positioning and direction of camera and object |
| CN110599507A (en) * | 2018-06-13 | 2019-12-20 | 中国农业大学 | Tomato identification and positioning method and system |
| CN110659694A (en) * | 2019-09-27 | 2020-01-07 | 华中农业大学 | Method for detecting citrus fruit base based on machine learning |
| CN111215345A (en) * | 2020-03-10 | 2020-06-02 | 虞结全 | Wheel hub outward appearance check out test set based on 3D vision |
| CN112101350A (en) * | 2020-09-03 | 2020-12-18 | 云南省计量测试技术研究院 | Empty box air pressure representation value image acquisition system, identification system and working method thereof |
| CN112857458A (en) * | 2021-02-05 | 2021-05-28 | 常州捷佳创精密机械有限公司 | Detection device, material platform equipment and image processing method thereof |
| CN112903241A (en) * | 2021-01-22 | 2021-06-04 | 湘潭大学 | Test system for simulating deep sea mining and operation method thereof |
| CN114170148A (en) * | 2021-11-12 | 2022-03-11 | 浙江托普云农科技股份有限公司 | Corn plant type parameter measuring method, system, device and storage medium |
| CN114638820A (en) * | 2022-03-30 | 2022-06-17 | 西北农林科技大学 | Plant phenotype monitoring device and method |
| CN115100507A (en) * | 2022-07-11 | 2022-09-23 | 郑州轻工业大学 | Intelligent identification system and method for air conditioner pipeline pointer instrument in dark environment |
| CN115631366A (en) * | 2022-09-29 | 2023-01-20 | 华南农业大学 | Tomato fruit maturity prediction method and device |
| CN115690452A (en) * | 2022-09-20 | 2023-02-03 | 山东易华录信息技术有限公司 | High-throughput fish phenotype analysis method and device based on machine vision |
-
2023
- 2023-06-02 CN CN202310649762.3A patent/CN116778474A/en active Pending
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE202005014072U1 (en) * | 2005-09-06 | 2005-11-24 | Maier, Florian | Apparatus photographing fixed objects from various perspectives, includes camera mounting plate on carriage and rails with computer control over relative positioning and direction of camera and object |
| CN110599507A (en) * | 2018-06-13 | 2019-12-20 | 中国农业大学 | Tomato identification and positioning method and system |
| CN110659694A (en) * | 2019-09-27 | 2020-01-07 | 华中农业大学 | Method for detecting citrus fruit base based on machine learning |
| CN111215345A (en) * | 2020-03-10 | 2020-06-02 | 虞结全 | Wheel hub outward appearance check out test set based on 3D vision |
| CN112101350A (en) * | 2020-09-03 | 2020-12-18 | 云南省计量测试技术研究院 | Empty box air pressure representation value image acquisition system, identification system and working method thereof |
| CN112903241A (en) * | 2021-01-22 | 2021-06-04 | 湘潭大学 | Test system for simulating deep sea mining and operation method thereof |
| CN112857458A (en) * | 2021-02-05 | 2021-05-28 | 常州捷佳创精密机械有限公司 | Detection device, material platform equipment and image processing method thereof |
| CN114170148A (en) * | 2021-11-12 | 2022-03-11 | 浙江托普云农科技股份有限公司 | Corn plant type parameter measuring method, system, device and storage medium |
| CN114638820A (en) * | 2022-03-30 | 2022-06-17 | 西北农林科技大学 | Plant phenotype monitoring device and method |
| CN115100507A (en) * | 2022-07-11 | 2022-09-23 | 郑州轻工业大学 | Intelligent identification system and method for air conditioner pipeline pointer instrument in dark environment |
| CN115690452A (en) * | 2022-09-20 | 2023-02-03 | 山东易华录信息技术有限公司 | High-throughput fish phenotype analysis method and device based on machine vision |
| CN115631366A (en) * | 2022-09-29 | 2023-01-20 | 华南农业大学 | Tomato fruit maturity prediction method and device |
Non-Patent Citations (4)
| Title |
|---|
| YIHANG ZHU等: "Quantitative Extraction and Evaluation of Tomato Fruit Phenotypes Based on Image Recognition", 《FRONTIER PLANT SCIENCE》, 13 April 2022 (2022-04-13), pages 1 - 12 * |
| 张寻梦等: "基于图像和YOLOv3 的番茄果实 表型参数计算及重量模拟", 《 江苏农业科学》, vol. 51, no. 10, 20 May 2023 (2023-05-20), pages 193 - 201 * |
| 王秀山等: "动态图像中烟株茎秆特征的识别与应用", 《烟草科技》, vol. 48, no. 10, 15 October 2015 (2015-10-15), pages 78 - 83 * |
| 贺语童: "基于深度神经网络的绿色作物分割模型研究", 《中国优秀硕士学位论文全文数据库 (农业科技辑)》, no. 2021, 15 December 2021 (2021-12-15), pages 043 - 16 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111443028B (en) | Automatic monitoring equipment and method for floating algae based on AI technology | |
| CN109887020B (en) | Plant organ separation method and system | |
| CN107944504B (en) | Board recognition and machine learning method and device for board recognition and electronic equipment | |
| CN111340798A (en) | Application of deep learning in product appearance flaw detection | |
| CN111968048B (en) | Method and system for enhancing image data of less power inspection samples | |
| CN112669348B (en) | Fish body posture estimation and fish body surface type data measurement method and device | |
| CN113743358B (en) | Landscape vision feature recognition method adopting omnibearing collection and intelligent calculation | |
| CN106846462A (en) | Insect identifying device and method based on three-dimensional simulation | |
| CN102819765A (en) | Milk somatic cell counting method based on computer vision | |
| CN117029673B (en) | Fish body surface multi-size measurement method based on artificial intelligence | |
| CN115355948A (en) | Method for detecting body size, body weight and backfat thickness of sow | |
| CN116682106A (en) | Deep learning-based intelligent detection method and device for diaphorina citri | |
| CN110619297B (en) | Bean fruiting body image batch acquisition and recognition method and device | |
| CN109919215B (en) | Target detection method for improving characteristic pyramid network based on clustering algorithm | |
| CN106886758A (en) | Based on insect identifying device and method that 3 d pose is estimated | |
| WO2022129362A1 (en) | A system for monitoring of dead fish | |
| CN116778474A (en) | Intelligent phenotype analyzer for tomato fruits | |
| CN109636856A (en) | Object 6 DOF degree posture information union measuring method based on HOG Fusion Features operator | |
| CN101021948A (en) | Automatic identifying device and method for joint in human body symmetric motion image | |
| CN108634241A (en) | Select halogen method and its brine-adding system | |
| CN113114766B (en) | Potted plant information detection method based on ZED camera | |
| CN113516635B (en) | Fish and vegetable symbiotic system and vegetable nitrogen element demand estimation method based on fish behaviors | |
| CN116646082A (en) | Disease early warning traceability system based on abnormal chicken manure under laminated cage raising mode | |
| CN107330938B (en) | Visual inspection method and system for multiple tested objects | |
| CN115876695A (en) | Sole rubber coating quality check out test set |
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 |