WO2020134255A1 - Procédé de surveillance de situations de croissance de poissons sur la base de la vision artificielle - Google Patents
Procédé de surveillance de situations de croissance de poissons sur la base de la vision artificielle Download PDFInfo
- Publication number
- WO2020134255A1 WO2020134255A1 PCT/CN2019/108334 CN2019108334W WO2020134255A1 WO 2020134255 A1 WO2020134255 A1 WO 2020134255A1 CN 2019108334 W CN2019108334 W CN 2019108334W WO 2020134255 A1 WO2020134255 A1 WO 2020134255A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- fish
- image
- pixel
- target
- gmm
- Prior art date
Links
- 241000251468 Actinopterygii Species 0.000 title claims abstract description 97
- 238000000034 method Methods 0.000 title claims abstract description 31
- 238000012544 monitoring process Methods 0.000 title claims abstract description 7
- 235000019688 fish Nutrition 0.000 title abstract description 68
- 230000006870 function Effects 0.000 claims description 13
- 230000011218 segmentation Effects 0.000 claims description 10
- 230000008569 process Effects 0.000 claims description 8
- 238000012417 linear regression Methods 0.000 claims description 7
- 238000012545 processing Methods 0.000 claims description 6
- 238000003709 image segmentation Methods 0.000 claims description 5
- 238000013528 artificial neural network Methods 0.000 claims description 4
- 230000005284 excitation Effects 0.000 claims description 4
- 238000000605 extraction Methods 0.000 claims description 4
- 238000007781 pre-processing Methods 0.000 claims description 4
- 238000013519 translation Methods 0.000 claims description 4
- 230000004913 activation Effects 0.000 claims description 2
- 238000004364 calculation method Methods 0.000 claims description 2
- 238000003384 imaging method Methods 0.000 claims description 2
- 238000010606 normalization Methods 0.000 claims description 2
- 230000010287 polarization Effects 0.000 claims description 2
- 238000005259 measurement Methods 0.000 abstract description 5
- 241000894007 species Species 0.000 abstract description 2
- 230000008901 benefit Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000009372 pisciculture Methods 0.000 description 3
- 238000009395 breeding Methods 0.000 description 2
- 230000001488 breeding effect Effects 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 241001465754 Metazoa Species 0.000 description 1
- 241000277269 Oncorhynchus masou Species 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 238000009313 farming Methods 0.000 description 1
- 230000006698 induction Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- CMJCEVKJYRZMIA-UHFFFAOYSA-M thallium(i) iodide Chemical compound [Tl]I CMJCEVKJYRZMIA-UHFFFAOYSA-M 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Images
Classifications
-
- A—HUMAN NECESSITIES
- A01—AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
- A01K—ANIMAL HUSBANDRY; AVICULTURE; APICULTURE; PISCICULTURE; FISHING; REARING OR BREEDING ANIMALS, NOT OTHERWISE PROVIDED FOR; NEW BREEDS OF ANIMALS
- A01K61/00—Culture of aquatic animals
- A01K61/90—Sorting, grading, counting or marking live aquatic animals, e.g. sex determination
- A01K61/95—Sorting, grading, counting or marking live aquatic animals, e.g. sex determination specially adapted for fish
-
- 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
Definitions
- the invention relates to underwater fish catching technology, in particular to a method for monitoring fish growth based on machine vision.
- the main fishing targets of marine fishery production are those densely populated groups of fish or economic animals that are migrating, breeding, bait or overwintering, especially breeding groups, which are dense and stable, and most fish are of the same body length Groups or groups of the same age group, such as salmon trout, are particularly noticeable.
- the technical problem to be solved by the present invention is to provide an underwater fishing method based on machine vision, which can catch fish of a specific species and a specific size, and at the same time, can also estimate the weight of fish and monitor the growth of fish in real time with precision Master the best fishing period.
- the present invention provides a method for monitoring the growth of fish based on machine vision, which is characterized by comprising: collecting an image of a fish using an underwater camera; according to the collected image of the fish Recognize and classify the fish; calculate the length of the fish based on the collected images of the fish; based on the result of the classification and the length of the fish, and based on the preset length and weight of the fish To estimate the weight of the fish.
- the advantage of the present invention is that the underwater fishing method based on machine vision of the present invention monitors the growth of fish in real time through the steps of fish identification and classification, fish length measurement and weight prediction through the underwater fishing device and accurately grasps the best Fishing period; that is, the length of the fish is measured by image processing to achieve accurate capture, the weight of the fish is estimated by image processing, and the growth of fish is monitored in real time to maximize the profit of farming, and then automatic fishing and analysis can be realized.
- the fish process simplifies a large number of fishing tasks and improves work efficiency.
- combining underwater robot technology and image recognition technology can help reduce the labor of personnel, increase production efficiency, and enhance the level of automation. It has theoretical and practical applications. significance.
- FIG. 1 is a flowchart of fish identification and classification in an underwater fishing method based on machine vision in an embodiment.
- FIG. 2 is a specific flowchart of image acquisition and preprocessing in FIG. 1.
- FIG. 3 is a flowchart of fish length measurement and weight prediction in an underwater fishing method based on machine vision in an embodiment.
- FIG. 4 is an intermediate image of fish body length prediction processing in an underwater fishing method based on machine vision in an embodiment.
- the invention provides a method for monitoring the growth of fish based on machine vision, which includes: using an underwater camera to collect fish images; identifying and classifying the fish according to the collected fish images; The collected fish image calculates the length of the fish; based on the results of the identification and classification and the length of the fish, and based on the preset relationship between the length and weight of the fish, the weight of the fish is estimated .
- the underwater fishing device in the underwater fishing method based on machine vision includes a Kangnai In-Sight7000 industrial camera for acquiring images, an underwater lighting lamp for dark environment lighting, and a fish induction
- the fish lure device and the underwater special fishing net and underwater robot for catching fish includes a three-color LED and a frequency conversion sounder; in addition, the blue-green light transmission under consideration is considered Offset, the underwater lighting is selected as a thallium iodide lamp, and the light energy it radiates is mostly concentrated in the blue and green range, and the water absorbs it very little; the underwater camera uses this light source compared with the incandescent lamp. Under the same power conditions, the efficiency is more than six times.
- the underwater fishing method based on machine vision monitors the growth of fish in real time through the underwater fishing device in conjunction with the steps of fish identification and classification, fish length measurement, and weight prediction to accurately grasp the best fishing period.
- the fish identification classification as shown in FIG. 1, specifically includes the following steps:
- Step 1 Image acquisition and preprocessing: As shown in Figure 2, the underwater camera collects color pictures of fish, filters the original image through an improved median filter, and then cuts the filtered picture into a rectangular image, and then The Grab Cut algorithm is used to segment the image to obtain a segmented image after removing the background, and then the grayscale, morphology and binarization operations are performed on the segmented image to obtain a two-dimensional binary image of the processed fish body;
- Step 2 Wavelet feature extraction:
- Step 3 BP neural network fish image classification
- Network initialization Use the moment feature of the target image obtained in the above steps as the input of the BP network to further identify the target; suppose the number of nodes in the input layer is n, the number of nodes in the hidden layer is l, and the number of nodes in the output layer The number is m, then the weight ⁇ ij from the input layer to the hidden layer, the weight ⁇ jk from the hidden layer to the output layer, the offset from the input layer to the hidden layer is a j , and the offset from the hidden layer to the output layer Is b k ; the learning rate is ⁇ and the excitation function is g(x); where the excitation function is g(x) and the Sigmoid function is taken in the form:
- the weight update formula is:
- the offset update formula is:
- the image segmentation performed in the above step 1 is to cut the filtered image into a rectangular image, and then use the Grab Cut algorithm to segment the image.
- the process of Grab Cut algorithm is as follows.
- Step 1.1 Color model:
- U is a regional item, which means that a pixel is classified as a target or background penalty, that is, the negative logarithm of the probability that a pixel belongs to the target or background.
- the mixed Gaussian density model is in the following form: And 0 ⁇ i ⁇ 1,
- Step 1.2 Iterative energy minimization segmentation algorithm:
- the Grab Cut algorithm is the smallest iteration, each iteration process makes the GMM modeling the target and background parameters better, and makes the image segmentation better.
- the iterative energy minimization segmentation algorithm includes:
- Step 1.2.1 Directly frame the target to obtain an initial trimap T, that is, the pixels outside the box are all used as background pixels TB, and the pixels in the box TU are all used as "may be target" pixels,
- GMM through the k-mean algorithm, the pixels belonging to the target and the background are clustered into K categories, that is, K Gaussian models in GMM.
- K Gaussian models in GMM Each Gaussian model in GMM has multiple pixel sample sets, which are estimated by its RGB value. Parameter mean and covariance, and the weight of the Gaussian component is determined by the ratio of the number of pixels belonging to the Gaussian component to the total number of pixels,
- Step 1.3 Iterative minimization:
- Step 1.3.1 Assign a Gaussian component in the GMM to each pixel, for example, pixel n is the target pixel, then substitute the RGB value of pixel n into each Gaussian component in the target GMM, the one with the highest probability is the most likely to generate n, that is, the k nth Gaussian component of pixel n:
- Step 1.3.2 For a given image data Z, learn to optimize the parameters of the GMM, which have been classified for each pixel into which Gaussian component in step 1.2, then each Gaussian model has more
- the parameter mean and covariance of the pixel samples are estimated by the RGB values of the pixel samples
- the weight of the Gaussian component is determined by the ratio of the number of pixels belonging to the Gaussian component to the total number of pixels:
- Step 1.3.3 Segmentation estimation: Through the Gibbs energy term analyzed in step 1.1, create a graph, and find the weights t-link and n-link, and then divide by the max flow/min cut algorithm:
- Step 1.3.4 Repeat steps 1.3.1 to 1.3.3 until convergence. After the segmentation of 3.3.3, whether each pixel belongs to the target GMM or the background GMM changes, so the k n of each pixel changes, so GMM Changes have also occurred, so each iteration will interactively optimize the GMM model and segmentation results. In addition, because the steps 1.3.1 to 1.3.3 are all energy-decreasing processes, it is guaranteed that the iteration process will converge.
- the fish length measurement and weight prediction as shown in FIG. 3, specifically include the following steps:
- Step 1 Obtaining fish sample length and weight parameters: By measuring the length and weight data of a large number of the same fish, and using linear regression to calculate the relationship between the two, and based on the final measured fish length Estimate fish weight, and then evaluate the growth of fish in the entire fishing ground, and whether it meets the capture conditions of underwater robots;
- Step 2 Extract fish length information: As shown in Figure 4, set a circle with a diameter of 5 cm at the end of the special underwater fishing net parallel to it, and make its imaging position in the upper left corner of the whole picture ; Based on the pre-processed image of the type identification, calculate the ratio of the number of pixels on the left and right of the picture after fish processing and the number of familiar points of the ring diameter multiplied by the ring diameter The length of the fish;
- Step 3 Length error compensation: the distance between the fish and the underwater special fishing net is 10-20cm. When calculating the length of the fish, add 5%-10% error compensation;
- Step 4 Weight prediction: The length information extracted from the preprocessed image is input into the linear regression function prediction model, and the approximate fish weight is calculated.
Landscapes
- Life Sciences & Earth Sciences (AREA)
- Zoology (AREA)
- Environmental Sciences (AREA)
- Engineering & Computer Science (AREA)
- Marine Sciences & Fisheries (AREA)
- Animal Husbandry (AREA)
- Biodiversity & Conservation Biology (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computational Linguistics (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- Artificial Intelligence (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Image Analysis (AREA)
Abstract
L'invention porte sur un procédé de surveillance de situations de croissance de poissons sur la base de la vision artificielle comprend les étapes consistant à : surveiller les situations de croissance de poissons en temps réel au moyen de la coopération d'un dispositif de pêche sous-marin et des étapes de reconnaissance et de classification des poissons, de la mesure de la longueur et la prédiction du poids du poisson, et de la maîtrise précise de la période de pêche optimale. Selon le procédé, il est possible de capturer des poissons d'espèces et de tailles spécifiques peuvent être pris, d'estimer le poids de poissons, de suivre en temps réel les situations de croissance de poissons, de maîtriser avec précision la période de pêche optimale.
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811608004.2 | 2018-12-27 | ||
CN201811608004.2A CN109784378A (zh) | 2018-12-27 | 2018-12-27 | 一种基于机器视觉的水下捕鱼方法 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2020134255A1 true WO2020134255A1 (fr) | 2020-07-02 |
Family
ID=66498475
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2019/108334 WO2020134255A1 (fr) | 2018-12-27 | 2019-09-27 | Procédé de surveillance de situations de croissance de poissons sur la base de la vision artificielle |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN109784378A (fr) |
WO (1) | WO2020134255A1 (fr) |
Cited By (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112419392A (zh) * | 2020-11-30 | 2021-02-26 | 广州博进信息技术有限公司 | 基于机器视觉计算运动物体实际大小的方法、设备及介质 |
CN113239324A (zh) * | 2021-04-13 | 2021-08-10 | 江苏农林职业技术学院 | 一种黑鱼性成熟度判别方法和判别系统 |
CN113591671A (zh) * | 2021-07-28 | 2021-11-02 | 常州大学 | 一种基于Mask-Rcnn识别鱼类生长检测方法 |
CN113628182A (zh) * | 2021-08-03 | 2021-11-09 | 中国农业大学 | 鱼类重量自动估计方法、装置、电子设备及存储介质 |
CN114241291A (zh) * | 2021-12-15 | 2022-03-25 | 广东省农业科学院动物科学研究所 | 基于惰性元素标记法与计算机视觉的诱食剂筛选方法 |
NO20201081A1 (en) * | 2020-10-05 | 2022-04-06 | Fishency Innovation As | Generating three dimensional skeleton representations of aquatic animals using machine learning |
CN114742806A (zh) * | 2022-04-21 | 2022-07-12 | 海南大学 | 一种基于关键点坐标回归的鱼体形态学特征测量方法 |
CN114998375A (zh) * | 2022-06-15 | 2022-09-02 | 中国农业大学 | 一种基于实例分割的活体鱼重量估计方法及系统 |
CN116843085A (zh) * | 2023-08-29 | 2023-10-03 | 深圳市明心数智科技有限公司 | 淡水鱼生长监控方法、装置、设备及存储介质 |
CN117011795A (zh) * | 2023-08-08 | 2023-11-07 | 南京农业大学 | 基于类高斯模糊支持度的河蟹生长状态无损监测评估平台及方法 |
CN117252885A (zh) * | 2023-09-05 | 2023-12-19 | 无锡科技职业学院 | 工业x光图像显著区域自动分割方法 |
CN118298289A (zh) * | 2024-06-06 | 2024-07-05 | 崂山国家实验室 | 基于多波束图像声呐的水下运动生物识别及评估方法 |
Families Citing this family (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109784378A (zh) * | 2018-12-27 | 2019-05-21 | 南京芊玥机器人科技有限公司 | 一种基于机器视觉的水下捕鱼方法 |
CN110378241B (zh) * | 2019-06-25 | 2022-04-29 | 北京百度网讯科技有限公司 | 作物生长状态监控方法、装置、计算机设备及存储介质 |
CN110347134A (zh) * | 2019-07-29 | 2019-10-18 | 南京图玩智能科技有限公司 | 一种ai智能水产养殖样本识别方法及养殖系统 |
CN111406693A (zh) * | 2020-04-23 | 2020-07-14 | 上海海洋大学 | 基于仿生海鳗的海洋牧场渔业资源养护效果评价方法 |
CN112287913B (zh) * | 2020-12-25 | 2021-04-06 | 浙江渔生泰科技有限公司 | 一种鱼类视频识别的智能监管系统 |
CN113205283B (zh) * | 2021-06-02 | 2023-09-12 | 中国水产科学研究院南海水产研究所 | 一种外海渔业捕捞调度方法及系统 |
CN114049577A (zh) * | 2021-11-17 | 2022-02-15 | 中国水利水电科学研究院 | 一种鱼类规格的测量方法及系统 |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103444613A (zh) * | 2013-08-29 | 2013-12-18 | 北京农业信息技术研究中心 | 一种用于鱼类养殖的投喂控制系统及方法 |
CN106561532A (zh) * | 2016-11-08 | 2017-04-19 | 深圳技师学院 | 一种监测鱼活动的方法和装置 |
CN107909137A (zh) * | 2017-11-27 | 2018-04-13 | 南瑞集团有限公司 | 一种鱼道过鱼计数和识别方法 |
WO2018111124A2 (fr) * | 2016-12-15 | 2018-06-21 | University Of The Philippines | Estimation d'une taille de poissons, d'une densité de population, d'une répartition d'espèces et d'une biomasse |
CN109784378A (zh) * | 2018-12-27 | 2019-05-21 | 南京芊玥机器人科技有限公司 | 一种基于机器视觉的水下捕鱼方法 |
-
2018
- 2018-12-27 CN CN201811608004.2A patent/CN109784378A/zh not_active Withdrawn
-
2019
- 2019-09-27 WO PCT/CN2019/108334 patent/WO2020134255A1/fr active Application Filing
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103444613A (zh) * | 2013-08-29 | 2013-12-18 | 北京农业信息技术研究中心 | 一种用于鱼类养殖的投喂控制系统及方法 |
CN106561532A (zh) * | 2016-11-08 | 2017-04-19 | 深圳技师学院 | 一种监测鱼活动的方法和装置 |
WO2018111124A2 (fr) * | 2016-12-15 | 2018-06-21 | University Of The Philippines | Estimation d'une taille de poissons, d'une densité de population, d'une répartition d'espèces et d'une biomasse |
CN107909137A (zh) * | 2017-11-27 | 2018-04-13 | 南瑞集团有限公司 | 一种鱼道过鱼计数和识别方法 |
CN109784378A (zh) * | 2018-12-27 | 2019-05-21 | 南京芊玥机器人科技有限公司 | 一种基于机器视觉的水下捕鱼方法 |
Cited By (19)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
NO20201081A1 (en) * | 2020-10-05 | 2022-04-06 | Fishency Innovation As | Generating three dimensional skeleton representations of aquatic animals using machine learning |
NO347281B1 (en) * | 2020-10-05 | 2023-08-21 | Fishency Innovation As | Generating three dimensional skeleton representations of aquatic animals using machine learning |
CN112419392A (zh) * | 2020-11-30 | 2021-02-26 | 广州博进信息技术有限公司 | 基于机器视觉计算运动物体实际大小的方法、设备及介质 |
CN113239324B (zh) * | 2021-04-13 | 2023-11-10 | 江苏农林职业技术学院 | 一种黑鱼性成熟度判别方法和判别系统 |
CN113239324A (zh) * | 2021-04-13 | 2021-08-10 | 江苏农林职业技术学院 | 一种黑鱼性成熟度判别方法和判别系统 |
CN113591671A (zh) * | 2021-07-28 | 2021-11-02 | 常州大学 | 一种基于Mask-Rcnn识别鱼类生长检测方法 |
CN113591671B (zh) * | 2021-07-28 | 2023-10-24 | 常州大学 | 一种基于Mask-Rcnn识别鱼类生长检测方法 |
CN113628182A (zh) * | 2021-08-03 | 2021-11-09 | 中国农业大学 | 鱼类重量自动估计方法、装置、电子设备及存储介质 |
CN113628182B (zh) * | 2021-08-03 | 2024-04-26 | 中国农业大学 | 鱼类重量自动估计方法、装置、电子设备及存储介质 |
CN114241291A (zh) * | 2021-12-15 | 2022-03-25 | 广东省农业科学院动物科学研究所 | 基于惰性元素标记法与计算机视觉的诱食剂筛选方法 |
CN114742806A (zh) * | 2022-04-21 | 2022-07-12 | 海南大学 | 一种基于关键点坐标回归的鱼体形态学特征测量方法 |
CN114742806B (zh) * | 2022-04-21 | 2024-10-01 | 海南大学 | 一种基于关键点坐标回归的鱼体形态学特征测量方法 |
CN114998375A (zh) * | 2022-06-15 | 2022-09-02 | 中国农业大学 | 一种基于实例分割的活体鱼重量估计方法及系统 |
CN117011795B (zh) * | 2023-08-08 | 2024-02-13 | 南京农业大学 | 基于类高斯模糊支持度的河蟹生长状态无损监测评估平台及方法 |
CN117011795A (zh) * | 2023-08-08 | 2023-11-07 | 南京农业大学 | 基于类高斯模糊支持度的河蟹生长状态无损监测评估平台及方法 |
CN116843085B (zh) * | 2023-08-29 | 2023-12-01 | 深圳市明心数智科技有限公司 | 淡水鱼生长监控方法、装置、设备及存储介质 |
CN116843085A (zh) * | 2023-08-29 | 2023-10-03 | 深圳市明心数智科技有限公司 | 淡水鱼生长监控方法、装置、设备及存储介质 |
CN117252885A (zh) * | 2023-09-05 | 2023-12-19 | 无锡科技职业学院 | 工业x光图像显著区域自动分割方法 |
CN118298289A (zh) * | 2024-06-06 | 2024-07-05 | 崂山国家实验室 | 基于多波束图像声呐的水下运动生物识别及评估方法 |
Also Published As
Publication number | Publication date |
---|---|
CN109784378A (zh) | 2019-05-21 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2020134255A1 (fr) | Procédé de surveillance de situations de croissance de poissons sur la base de la vision artificielle | |
Yang et al. | Computer vision models in intelligent aquaculture with emphasis on fish detection and behavior analysis: a review | |
Zhou et al. | Evaluation of fish feeding intensity in aquaculture using a convolutional neural network and machine vision | |
Mandal et al. | Assessing fish abundance from underwater video using deep neural networks | |
US11636701B2 (en) | Method for calculating deviation relations of a population | |
CN112598713A (zh) | 一种基于深度学习的近岸海底鱼类检测、跟踪统计方法 | |
CN109684906B (zh) | 基于深度学习的检测红脂大小蠹的方法 | |
CN111406693A (zh) | 基于仿生海鳗的海洋牧场渔业资源养护效果评价方法 | |
Lainez et al. | Automated fingerlings counting using convolutional neural network | |
CN113349111A (zh) | 一种用于水产养殖物的动态投喂方法、系统和存储介质 | |
Almero et al. | An aquaculture-based binary classifier for fish detection using multilayer artificial neural network | |
Cao et al. | Automatic coarse-to-fine joint detection and segmentation of underwater non-structural live crabs for precise feeding | |
CN115578423A (zh) | 基于深度学习的鱼类关键点检测、个体追踪和生物量估算方法及系统 | |
Lai et al. | Automatic measuring shrimp body length using CNN and an underwater imaging system | |
CN115410152A (zh) | 一种户外鱼类池塘养殖投喂监控系统 | |
Sosa-Trejo et al. | Vision-based techniques for automatic marine plankton classification | |
Xu et al. | Detection of bluefin tuna by cascade classifier and deep learning for monitoring fish resources | |
CN113627558A (zh) | 鱼类图像识别方法、系统及设备 | |
CN113284164A (zh) | 虾群自动计数方法、装置、电子设备及存储介质 | |
Ge et al. | Real-time object detection algorithm for Underwater Robots | |
CN112308002B (zh) | 一种基于单阶段深度学习网络的海底生物识别和检测方法 | |
Hassoon | Fish species identification techniques: a review | |
Kaveti et al. | Towards automated fish detection using convolutional neural networks | |
Kim et al. | Recognition of adherent polychaetes on oysters and scallops using Microsoft Azure Custom Vision | |
TW202309823A (zh) | 人工智慧養殖魚群食慾判斷方法與系統 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 19901657 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 19901657 Country of ref document: EP Kind code of ref document: A1 |