Pub Date : 2025-02-06DOI: 10.1016/j.aiia.2025.01.012
Hang Yang , Qi Feng , Shibin Xia , Zhenbin Wu , Yi Zhang
The integration of artificial intelligence (AI) in aquaculture has been identified as a transformative force, enhancing various operational aspects from water quality management to genetic optimization. This review provides a comprehensive synthesis of recent advancements in AI applications within the aquaculture sector, underscoring the significant enhancements in production efficiency and environmental sustainability. Key AI-driven improvements, such as predictive analytics for disease management and optimized feeding protocols, are highlighted, demonstrating their contributions to reducing waste and improving biomass outputs. However, challenges remain in terms of data quality, system integration, and the socio-economic impacts of technological adoption across diverse aquacultural environments. This review also addresses the gaps in current research, particularly the lack of robust, scalable AI models and frameworks that can be universally applied. Future directions are discussed, emphasizing the need for interdisciplinary research and development to fully leverage AI potential in aquaculture. This study not only maps the current landscape of AI applications but also serves as a call for continued innovation and strategic collaborations to overcome existing barriers and realize the full benefits of AI in aquaculture.
{"title":"AI-driven aquaculture: A review of technological innovations and their sustainable impacts","authors":"Hang Yang , Qi Feng , Shibin Xia , Zhenbin Wu , Yi Zhang","doi":"10.1016/j.aiia.2025.01.012","DOIUrl":"10.1016/j.aiia.2025.01.012","url":null,"abstract":"<div><div>The integration of artificial intelligence (AI) in aquaculture has been identified as a transformative force, enhancing various operational aspects from water quality management to genetic optimization. This review provides a comprehensive synthesis of recent advancements in AI applications within the aquaculture sector, underscoring the significant enhancements in production efficiency and environmental sustainability. Key AI-driven improvements, such as predictive analytics for disease management and optimized feeding protocols, are highlighted, demonstrating their contributions to reducing waste and improving biomass outputs. However, challenges remain in terms of data quality, system integration, and the socio-economic impacts of technological adoption across diverse aquacultural environments. This review also addresses the gaps in current research, particularly the lack of robust, scalable AI models and frameworks that can be universally applied. Future directions are discussed, emphasizing the need for interdisciplinary research and development to fully leverage AI potential in aquaculture. This study not only maps the current landscape of AI applications but also serves as a call for continued innovation and strategic collaborations to overcome existing barriers and realize the full benefits of AI in aquaculture.</div></div>","PeriodicalId":52814,"journal":{"name":"Artificial Intelligence in Agriculture","volume":"15 3","pages":"Pages 508-525"},"PeriodicalIF":8.2,"publicationDate":"2025-02-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144106613","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-02-03DOI: 10.1016/j.aiia.2025.01.008
Anthony Crespo , Claudia Moncada , Fabricio Crespo , Manuel Eugenio Morocho-Cayamcela
Currently, artificial intelligence (AI), particularly computer vision (CV), has numerous applications in agriculture. In this field, the production and consumption of strawberries have experienced great growth in recent years, which makes meeting the growing demand a challenge that producers must face. However, one of the main problems regarding the cultivation of this fruit is the high cost and long picking times. In response, automatic harvesting has surged as an option to address this difficulty, and fruit instance segmentation plays a crucial role in these types of systems. Fruit segmentation is related to the identification and separation of individual fruits within a crop, allowing a more efficient and accurate harvesting process. Although deep learning (DL) techniques have shown potential for this activity, the complexity of the models leads to difficulty in their implementation in real-time systems. For this reason, a model capable of performing adequately in real-time, while also having good precision is of great interest. With this motivation, this work presents a efficient Mask R-CNN model to perform instance segmentation in strawberry fruits. The efficiency of the model is assessed considering the amount of frames per second (FPS) it can process, its size in megabytes (MB) and its mean average precision (mAP) value. Two approaches are provided: The first one consists on the training of the model using the Detectron2 library, while the second one focuses on the training of the model using the NVIDIA TAO Toolkit. In both cases, NVIDIA TensorRT is used to optimize the models. The results show that the best Mask R-CNN model, without optimization, has a performance of 83.45 mAP, 4 FPS, and 351 MB of size, which, after the TensorRT optimization, achieved 83.17 mAP, 25.46 FPS, and only 48.2 MB of size. It represents a suitable model for implementation in real-time systems.
{"title":"An efficient strawberry segmentation model based on Mask R-CNN and TensorRT","authors":"Anthony Crespo , Claudia Moncada , Fabricio Crespo , Manuel Eugenio Morocho-Cayamcela","doi":"10.1016/j.aiia.2025.01.008","DOIUrl":"10.1016/j.aiia.2025.01.008","url":null,"abstract":"<div><div>Currently, artificial intelligence (AI), particularly computer vision (CV), has numerous applications in agriculture. In this field, the production and consumption of strawberries have experienced great growth in recent years, which makes meeting the growing demand a challenge that producers must face. However, one of the main problems regarding the cultivation of this fruit is the high cost and long picking times. In response, automatic harvesting has surged as an option to address this difficulty, and fruit instance segmentation plays a crucial role in these types of systems. Fruit segmentation is related to the identification and separation of individual fruits within a crop, allowing a more efficient and accurate harvesting process. Although deep learning (DL) techniques have shown potential for this activity, the complexity of the models leads to difficulty in their implementation in real-time systems. For this reason, a model capable of performing adequately in real-time, while also having good precision is of great interest. With this motivation, this work presents a efficient Mask R-CNN model to perform instance segmentation in strawberry fruits. The efficiency of the model is assessed considering the amount of frames per second (FPS) it can process, its size in megabytes (MB) and its mean average precision (mAP) value. Two approaches are provided: The first one consists on the training of the model using the Detectron2 library, while the second one focuses on the training of the model using the NVIDIA TAO Toolkit. In both cases, NVIDIA TensorRT is used to optimize the models. The results show that the best Mask R-CNN model, without optimization, has a performance of 83.45 mAP, 4 FPS, and 351 MB of size, which, after the TensorRT optimization, achieved 83.17 mAP, 25.46 FPS, and only 48.2 MB of size. It represents a suitable model for implementation in real-time systems.</div></div>","PeriodicalId":52814,"journal":{"name":"Artificial Intelligence in Agriculture","volume":"15 2","pages":"Pages 327-337"},"PeriodicalIF":8.2,"publicationDate":"2025-02-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143704966","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-01-27DOI: 10.1016/j.aiia.2025.01.011
Leng Han , Zhichong Wang , Miao He , Yajia Liu , Xiongkui He
Precision application in orchards enhancing deposition uniformity and environmental sustainability by accurately matching nozzle output with canopy parameters. This study provides a pipeline for creating 3D prescription maps using a UAV and performing offline variable application. It also evaluates the accuracy of ground altitude measurements at various flight heights. At a flight height of 30 m, with a three-dimensional reconstruction method without phase-control points, the root mean square error (RMSE) for ground altitude measurement was 0.214 m and the mean absolute error (MAE) was 0.211 m; for the canopy area, these values were 0.591 m and 0.541 m, respectively. As flight height increased, the accuracy of altitude measurements declined and tended to be underestimated. Moreover, during offline variable spraying, the shape of the spray area influenced deposition accuracy, with collision detection area of a line segment achieving greater precision than conical ones. Field tests showed that the offline variable application method reduced pesticide usage by 32.43 % and enhanced spray uniformity. This newly developed process does not require costly sensors on each sprayer and has potential for field applications.
{"title":"PWM offline variable application based on UAV remote sensing 3D prescription map","authors":"Leng Han , Zhichong Wang , Miao He , Yajia Liu , Xiongkui He","doi":"10.1016/j.aiia.2025.01.011","DOIUrl":"10.1016/j.aiia.2025.01.011","url":null,"abstract":"<div><div>Precision application in orchards enhancing deposition uniformity and environmental sustainability by accurately matching nozzle output with canopy parameters. This study provides a pipeline for creating 3D prescription maps using a UAV and performing offline variable application. It also evaluates the accuracy of ground altitude measurements at various flight heights. At a flight height of 30 m, with a three-dimensional reconstruction method without phase-control points, the root mean square error (RMSE) for ground altitude measurement was 0.214 m and the mean absolute error (MAE) was 0.211 m; for the canopy area, these values were 0.591 m and 0.541 m, respectively. As flight height increased, the accuracy of altitude measurements declined and tended to be underestimated. Moreover, during offline variable spraying, the shape of the spray area influenced deposition accuracy, with collision detection area of a line segment achieving greater precision than conical ones. Field tests showed that the offline variable application method reduced pesticide usage by 32.43 % and enhanced spray uniformity. This newly developed process does not require costly sensors on each sprayer and has potential for field applications.</div></div>","PeriodicalId":52814,"journal":{"name":"Artificial Intelligence in Agriculture","volume":"15 3","pages":"Pages 496-507"},"PeriodicalIF":8.2,"publicationDate":"2025-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144106612","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The swift evolution of deep learning has greatly benefited the field of intensive aquaculture. Specifically, deep learning-based shrimp larvae detection has offered important technical assistance for counting shrimp larvae and recognizing abnormal behaviors. Firstly, the transparent bodies and small sizes of shrimp larvae, combined with complex scenarios due to variations in light intensity and water turbidity, make it challenging for current detection methods to achieve high accuracy. Secondly, deep learning-based object detection demands substantial computing power and storage space, which restricts its application on edge devices. This paper proposes an efficient one-stage shrimp larvae detection method, FAMDet, specifically designed for complex scenarios in intensive aquaculture. Firstly, different from the ordinary detection methods, it exploits an efficient FasterNet backbone, constructed with partial convolution, to extract effective multi-scale shrimp larvae features. Meanwhile, we construct an adaptively bi-directional fusion neck to integrate high-level semantic information and low-level detail information of shrimp larvae in a matter that sufficiently merges features and further mitigates noise interference. Finally, a decoupled detection head equipped with MPDIoU is used for precise bounding box regression of shrimp larvae. We collected images of shrimp larvae from multiple scenarios and labeled 108,365 targets for experiments. Compared with the ordinary detection methods (Faster RCNN, SSD, RetinaNet, CenterNet, FCOS, DETR, and YOLOX_s), FAMDet has obtained considerable advantages in accuracy, speed, and complexity. Compared with the outstanding one-stage method YOLOv8s, it has improved accuracy while reducing 57 % parameters, 37 % FLOPs, 22 % inference latency per image on CPU, and 56 % storage overhead. Furthermore, FAMDet has still outperformed multiple lightweight methods (EfficientDet, RT-DETR, GhostNetV2, EfficientFormerV2, EfficientViT, and MobileNetV4). In addition, we conducted experiments on the public dataset (VOC 07 + 12) to further verify the effectiveness of FAMDet. Consequently, the proposed method can effectively alleviate the limitations faced by resource-constrained devices and achieve superior shrimp larvae detection results.
{"title":"Efficient one-stage detection of shrimp larvae in complex aquaculture scenarios","authors":"Guoxu Zhang , Tianyi Liao , Yingyi Chen , Ping Zhong , Zhencai Shen , Daoliang Li","doi":"10.1016/j.aiia.2025.01.009","DOIUrl":"10.1016/j.aiia.2025.01.009","url":null,"abstract":"<div><div>The swift evolution of deep learning has greatly benefited the field of intensive aquaculture. Specifically, deep learning-based shrimp larvae detection has offered important technical assistance for counting shrimp larvae and recognizing abnormal behaviors. Firstly, the transparent bodies and small sizes of shrimp larvae, combined with complex scenarios due to variations in light intensity and water turbidity, make it challenging for current detection methods to achieve high accuracy. Secondly, deep learning-based object detection demands substantial computing power and storage space, which restricts its application on edge devices. This paper proposes an efficient one-stage shrimp larvae detection method, FAMDet, specifically designed for complex scenarios in intensive aquaculture. Firstly, different from the ordinary detection methods, it exploits an efficient FasterNet backbone, constructed with partial convolution, to extract effective multi-scale shrimp larvae features. Meanwhile, we construct an adaptively bi-directional fusion neck to integrate high-level semantic information and low-level detail information of shrimp larvae in a matter that sufficiently merges features and further mitigates noise interference. Finally, a decoupled detection head equipped with MPDIoU is used for precise bounding box regression of shrimp larvae. We collected images of shrimp larvae from multiple scenarios and labeled 108,365 targets for experiments. Compared with the ordinary detection methods (Faster RCNN, SSD, RetinaNet, CenterNet, FCOS, DETR, and YOLOX_s), FAMDet has obtained considerable advantages in accuracy, speed, and complexity. Compared with the outstanding one-stage method YOLOv8s, it has improved accuracy while reducing 57 % parameters, 37 % FLOPs, 22 % inference latency per image on CPU, and 56 % storage overhead. Furthermore, FAMDet has still outperformed multiple lightweight methods (EfficientDet, RT-DETR, GhostNetV2, EfficientFormerV2, EfficientViT, and MobileNetV4). In addition, we conducted experiments on the public dataset (VOC 07 + 12) to further verify the effectiveness of FAMDet. Consequently, the proposed method can effectively alleviate the limitations faced by resource-constrained devices and achieve superior shrimp larvae detection results.</div></div>","PeriodicalId":52814,"journal":{"name":"Artificial Intelligence in Agriculture","volume":"15 2","pages":"Pages 338-349"},"PeriodicalIF":8.2,"publicationDate":"2025-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143704748","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-01-27DOI: 10.1016/j.aiia.2025.01.010
Jingwen Li , Pengbo Zeng , Shuai Yue , Zhiyang Zheng , Lifeng Qin , Huaibo Song
This study proposes an automated scoring system for cow body condition using improved YOLOv5 to assess the body condition distribution of herd cows, which significantly impacts herd productivity and feeding management. A dataset was created by capturing images of the cow's hindquarters using an image sensor at the entrance of the milking hall. This system enhances feature extraction ability by introducing dual path networks and convolutional block attention modules and improves efficiency by replacing some modules from the standard YOLOv5s with deep separable convolution to reduce parameters. Furthermore, the system employs an automatic detection and segmentation algorithm to achieve individual cow segmentation and body condition acquisition in the video. Subsequently, the system computes the body condition distribution of cows in a group state. The experimental findings demonstrate that the proposed model outperforms the original YOLOv5 network with higher accuracy and fewer computations and parameters. The precision, recall, and mean average precision of the model are 94.3 %, 92.5 %, and 91.8 %, respectively. The algorithm achieved an overall detection rate of 94.2 % for individual cow segmentation and body condition acquisition in the video, with a body condition scoring accuracy of 92.5 % among accurately detected cows and an overall body condition scoring accuracy of 87.1 % across the 10 video tests.
{"title":"Automatic body condition scoring system for dairy cows in group state based on improved YOLOv5 and video analysis","authors":"Jingwen Li , Pengbo Zeng , Shuai Yue , Zhiyang Zheng , Lifeng Qin , Huaibo Song","doi":"10.1016/j.aiia.2025.01.010","DOIUrl":"10.1016/j.aiia.2025.01.010","url":null,"abstract":"<div><div>This study proposes an automated scoring system for cow body condition using improved YOLOv5 to assess the body condition distribution of herd cows, which significantly impacts herd productivity and feeding management. A dataset was created by capturing images of the cow's hindquarters using an image sensor at the entrance of the milking hall. This system enhances feature extraction ability by introducing dual path networks and convolutional block attention modules and improves efficiency by replacing some modules from the standard YOLOv5s with deep separable convolution to reduce parameters. Furthermore, the system employs an automatic detection and segmentation algorithm to achieve individual cow segmentation and body condition acquisition in the video. Subsequently, the system computes the body condition distribution of cows in a group state. The experimental findings demonstrate that the proposed model outperforms the original YOLOv5 network with higher accuracy and fewer computations and parameters. The precision, recall, and mean average precision of the model are 94.3 %, 92.5 %, and 91.8 %, respectively. The algorithm achieved an overall detection rate of 94.2 % for individual cow segmentation and body condition acquisition in the video, with a body condition scoring accuracy of 92.5 % among accurately detected cows and an overall body condition scoring accuracy of 87.1 % across the 10 video tests.</div></div>","PeriodicalId":52814,"journal":{"name":"Artificial Intelligence in Agriculture","volume":"15 2","pages":"Pages 350-362"},"PeriodicalIF":8.2,"publicationDate":"2025-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143705546","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Climate change has intensified maize stalk lodging, severely impacting global maize production. While numerous traits influence stalk lodging resistance, their relative importance remains unclear, hindering breeding efforts. This study introduces an combining wind tunnel testing with machine learning algorithms to quantitatively evaluate stalk lodging resistance traits. Through extensive field experiments and literature review, we identified and measured 74 phenotypic traits encompassing plant morphology, biomass, and anatomical characteristics in maize plants. Correlation analysis revealed a median linear correlation coefficient of 0.497 among these traits, with 15.1 % of correlations exceeding 0.8. Principal component analysis showed that the first five components explained 90 % of the total variance, indicating significant trait interactions. Through feature engineering and gradient boosting regression, we developed a high-precision wind speed-ear displacement prediction model (R2 = 0.93) and identified 29 key traits critical for stalk lodging resistance. Sensitivity analysis revealed plant height as the most influential factor (sensitivity coefficient: −3.87), followed by traits of the 7th internode including epidermis layer thickness (0.62), pith area (−0.60), and lignin content (0.35). Our methodological framework not only provides quantitative insights into maize stalk lodging resistance mechanisms but also establishes a systematic approach for trait evaluation. The findings offer practical guidance for breeding programs focused on enhancing stalk lodging resistance and yield stability under climate change conditions, with potential applications in agronomic practice optimization and breeding strategy development.
{"title":"Identifying key factors influencing maize stalk lodging resistance through wind tunnel simulations with machine learning algorithms","authors":"Guanmin Huang, Ying Zhang, Shenghao Gu, Weiliang Wen, Xianju Lu, Xinyu Guo","doi":"10.1016/j.aiia.2025.01.007","DOIUrl":"10.1016/j.aiia.2025.01.007","url":null,"abstract":"<div><div>Climate change has intensified maize stalk lodging, severely impacting global maize production. While numerous traits influence stalk lodging resistance, their relative importance remains unclear, hindering breeding efforts. This study introduces an combining wind tunnel testing with machine learning algorithms to quantitatively evaluate stalk lodging resistance traits. Through extensive field experiments and literature review, we identified and measured 74 phenotypic traits encompassing plant morphology, biomass, and anatomical characteristics in maize plants. Correlation analysis revealed a median linear correlation coefficient of 0.497 among these traits, with 15.1 % of correlations exceeding 0.8. Principal component analysis showed that the first five components explained 90 % of the total variance, indicating significant trait interactions. Through feature engineering and gradient boosting regression, we developed a high-precision wind speed-ear displacement prediction model (R<sup>2</sup> = 0.93) and identified 29 key traits critical for stalk lodging resistance. Sensitivity analysis revealed plant height as the most influential factor (sensitivity coefficient: −3.87), followed by traits of the 7th internode including epidermis layer thickness (0.62), pith area (−0.60), and lignin content (0.35). Our methodological framework not only provides quantitative insights into maize stalk lodging resistance mechanisms but also establishes a systematic approach for trait evaluation. The findings offer practical guidance for breeding programs focused on enhancing stalk lodging resistance and yield stability under climate change conditions, with potential applications in agronomic practice optimization and breeding strategy development.</div></div>","PeriodicalId":52814,"journal":{"name":"Artificial Intelligence in Agriculture","volume":"15 2","pages":"Pages 316-326"},"PeriodicalIF":8.2,"publicationDate":"2025-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143704965","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-01-11DOI: 10.1016/j.aiia.2025.01.006
Hongli Song , Weiliang Wen , Sheng Wu , Xinyu Guo
Segmentation of three-dimensional (3D) point clouds is fundamental in comprehending unstructured structural and morphological data. It plays a critical role in research related to plant phenomics, 3D plant modeling, and functional-structural plant modeling. Although technologies for plant point cloud segmentation (PPCS) have advanced rapidly, there has been a lack of a systematic overview of the development process. This paper presents an overview of the progress made in 3D point cloud segmentation research in plants. It starts by discussing the methods used to acquire point clouds in plants, and analyzes the impact of point cloud resolution and quality on the segmentation task. It then introduces multi-scale point cloud segmentation in plants. The paper summarizes and analyzes traditional methods for PPCS, including the global and local features. This paper discusses the progress of machine learning-based segmentation on plant point clouds through supervised, unsupervised, and integrated approaches. It also summarizes the datasets that for PPCS using deep learning-oriented methods and explains the advantages and disadvantages of deep learning-based methods for projection-based, voxel-based, and point-based approaches respectively. Finally, the development of PPCS is discussed and prospected. Deep learning methods are predicted to become dominant in the field of PPCS, and 3D point cloud segmentation would develop towards more automated with higher resolution and precision.
{"title":"Comprehensive review on 3D point cloud segmentation in plants","authors":"Hongli Song , Weiliang Wen , Sheng Wu , Xinyu Guo","doi":"10.1016/j.aiia.2025.01.006","DOIUrl":"10.1016/j.aiia.2025.01.006","url":null,"abstract":"<div><div>Segmentation of three-dimensional (3D) point clouds is fundamental in comprehending unstructured structural and morphological data. It plays a critical role in research related to plant phenomics, 3D plant modeling, and functional-structural plant modeling. Although technologies for plant point cloud segmentation (PPCS) have advanced rapidly, there has been a lack of a systematic overview of the development process. This paper presents an overview of the progress made in 3D point cloud segmentation research in plants. It starts by discussing the methods used to acquire point clouds in plants, and analyzes the impact of point cloud resolution and quality on the segmentation task. It then introduces multi-scale point cloud segmentation in plants. The paper summarizes and analyzes traditional methods for PPCS, including the global and local features. This paper discusses the progress of machine learning-based segmentation on plant point clouds through supervised, unsupervised, and integrated approaches. It also summarizes the datasets that for PPCS using deep learning-oriented methods and explains the advantages and disadvantages of deep learning-based methods for projection-based, voxel-based, and point-based approaches respectively. Finally, the development of PPCS is discussed and prospected. Deep learning methods are predicted to become dominant in the field of PPCS, and 3D point cloud segmentation would develop towards more automated with higher resolution and precision.</div></div>","PeriodicalId":52814,"journal":{"name":"Artificial Intelligence in Agriculture","volume":"15 2","pages":"Pages 296-315"},"PeriodicalIF":8.2,"publicationDate":"2025-01-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143704964","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-01-10DOI: 10.1016/j.aiia.2025.01.003
Tao Cheng , Dongyan Zhang , Gan Zhang , Tianyi Wang , Weibo Ren , Feng Yuan , Yaling Liu , Zhaoming Wang , Chunjiang Zhao
High-throughput phenotyping (HTP) technology is now a significant bottleneck in the efficient selection and breeding of superior forage genetic resources. To better understand the status of forage phenotyping research and identify key directions for development, this review summarizes advances in HTP technology for forage phenotypic analysis over the past ten years. This paper reviews the unique aspects and research priorities in forage phenotypic monitoring, highlights key remote sensing platforms, examines the applications of advanced sensing technology for quantifying phenotypic traits, explores artificial intelligence (AI) algorithms in phenotypic data integration and analysis, and assesses recent progress in phenotypic genomics. The practical applications of HTP technology in forage remain constrained by several challenges. These include establishing uniform data collection standards, designing effective algorithms to handle complex genetic and environmental interactions, deepening the cross-exploration of phenomics-genomics, solving the problem of pathological inversion of forage phenotypic growth monitoring models, and developing low-cost forage phenotypic equipment. Resolving these challenges will unlock the full potential of HTP, enabling precise identification of superior forage traits, accelerating the breeding of superior varieties, and ultimately improving forage yield.
{"title":"High-throughput phenotyping techniques for forage: Status, bottleneck, and challenges","authors":"Tao Cheng , Dongyan Zhang , Gan Zhang , Tianyi Wang , Weibo Ren , Feng Yuan , Yaling Liu , Zhaoming Wang , Chunjiang Zhao","doi":"10.1016/j.aiia.2025.01.003","DOIUrl":"10.1016/j.aiia.2025.01.003","url":null,"abstract":"<div><div>High-throughput phenotyping (HTP) technology is now a significant bottleneck in the efficient selection and breeding of superior forage genetic resources. To better understand the status of forage phenotyping research and identify key directions for development, this review summarizes advances in HTP technology for forage phenotypic analysis over the past ten years. This paper reviews the unique aspects and research priorities in forage phenotypic monitoring, highlights key remote sensing platforms, examines the applications of advanced sensing technology for quantifying phenotypic traits, explores artificial intelligence (AI) algorithms in phenotypic data integration and analysis, and assesses recent progress in phenotypic genomics. The practical applications of HTP technology in forage remain constrained by several challenges. These include establishing uniform data collection standards, designing effective algorithms to handle complex genetic and environmental interactions, deepening the cross-exploration of phenomics-genomics, solving the problem of pathological inversion of forage phenotypic growth monitoring models, and developing low-cost forage phenotypic equipment. Resolving these challenges will unlock the full potential of HTP, enabling precise identification of superior forage traits, accelerating the breeding of superior varieties, and ultimately improving forage yield.</div></div>","PeriodicalId":52814,"journal":{"name":"Artificial Intelligence in Agriculture","volume":"15 1","pages":"Pages 98-115"},"PeriodicalIF":8.2,"publicationDate":"2025-01-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143097558","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-01-10DOI: 10.1016/j.aiia.2025.01.002
Artzai Picon , Itziar Eguskiza , Pablo Galan , Laura Gomez-Zamanillo , Javier Romero , Christian Klukas , Arantza Bereciartua-Perez , Mike Scharner , Ramon Navarra-Mestre
In this study, we introduced an innovative crop-conditional semantic segmentation architecture that seamlessly incorporates contextual metadata (crop information). This is achieved by merging the contextual information at a late layer stage, allowing the method to be integrated with any semantic segmentation architecture, including novel ones. To evaluate the effectiveness of this approach, we curated a challenging dataset of over 100,000 images captured in real-field conditions using mobile phones. This dataset includes various disease stages across 21 diseases and seven crops (wheat, barley, corn, rice, rape-seed, vinegrape, and cucumber), with the added complexity of multiple diseases coexisting in a single image. We demonstrate that incorporating contextual multi-crop information significantly enhances the performance of semantic segmentation models for plant disease detection. By leveraging crop-specific metadata, our approach achieves higher accuracy and better generalization across diverse crops (F1 = 0.68, r = 0.75) compared to traditional methods (F1 = 0.24, r = 0.68). Additionally, the adoption of a semi-supervised approach based on pseudo-labeling of single diseased plants, offers significant advantages for plant disease segmentation and quantification (F1 = 0.73, r = 0.95). This method enhances the model's performance by leveraging both labeled and unlabeled data, reducing the dependency on extensive manual annotations, which are often time-consuming and costly.
The deployment of this algorithm holds the potential to revolutionize the digitization of crop protection product testing, ensuring heightened repeatability while minimizing human subjectivity. By addressing the challenges of semantic segmentation and disease quantification, we contribute to more effective and precise phenotyping, ultimately supporting better crop management and protection strategies.
在这项研究中，我们引入了一种创新的作物条件语义分割架构，该架构无缝地结合了上下文元数据（作物信息）。这是通过在后期阶段合并上下文信息来实现的，允许该方法与任何语义分割体系结构集成，包括新的。为了评估这种方法的有效性，我们策划了一个具有挑战性的数据集，其中包括使用手机在实际条件下拍摄的100,000多张图像。该数据集包括21种疾病和7种作物（小麦、大麦、玉米、水稻、油菜籽、葡萄和黄瓜）的不同疾病阶段，并且在单个图像中同时存在多种疾病的复杂性。我们证明，结合上下文多作物信息显着提高了植物病害检测的语义分割模型的性能。通过利用特定作物的元数据，与传统方法（F1 = 0.24, r = 0.68）相比，我们的方法在不同作物之间实现了更高的精度和更好的泛化（F1 = 0.68, r = 0.75）。此外，采用基于单株病株伪标记的半监督方法，对植物病害的分割和定量具有显著优势（F1 = 0.73, r = 0.95）。该方法通过利用标记和未标记的数据来增强模型的性能，减少了对大量手工注释的依赖，而手工注释通常既耗时又昂贵。该算法的部署有可能彻底改变作物保护产品测试的数字化，确保提高可重复性，同时最大限度地减少人类的主观性。通过解决语义分割和疾病量化的挑战，我们有助于更有效和精确的表型分析，最终支持更好的作物管理和保护策略。
{"title":"Crop-conditional semantic segmentation for efficient agricultural disease assessment","authors":"Artzai Picon , Itziar Eguskiza , Pablo Galan , Laura Gomez-Zamanillo , Javier Romero , Christian Klukas , Arantza Bereciartua-Perez , Mike Scharner , Ramon Navarra-Mestre","doi":"10.1016/j.aiia.2025.01.002","DOIUrl":"10.1016/j.aiia.2025.01.002","url":null,"abstract":"<div><div>In this study, we introduced an innovative crop-conditional semantic segmentation architecture that seamlessly incorporates contextual metadata (crop information). This is achieved by merging the contextual information at a late layer stage, allowing the method to be integrated with any semantic segmentation architecture, including novel ones. To evaluate the effectiveness of this approach, we curated a challenging dataset of over 100,000 images captured in real-field conditions using mobile phones. This dataset includes various disease stages across 21 diseases and seven crops (wheat, barley, corn, rice, rape-seed, vinegrape, and cucumber), with the added complexity of multiple diseases coexisting in a single image. We demonstrate that incorporating contextual multi-crop information significantly enhances the performance of semantic segmentation models for plant disease detection. By leveraging crop-specific metadata, our approach achieves higher accuracy and better generalization across diverse crops (F1 = 0.68, <em>r</em> = 0.75) compared to traditional methods (F1 = 0.24, <em>r</em> = 0.68). Additionally, the adoption of a semi-supervised approach based on pseudo-labeling of single diseased plants, offers significant advantages for plant disease segmentation and quantification (F1 = 0.73, <em>r</em> = 0.95). This method enhances the model's performance by leveraging both labeled and unlabeled data, reducing the dependency on extensive manual annotations, which are often time-consuming and costly.</div><div>The deployment of this algorithm holds the potential to revolutionize the digitization of crop protection product testing, ensuring heightened repeatability while minimizing human subjectivity. By addressing the challenges of semantic segmentation and disease quantification, we contribute to more effective and precise phenotyping, ultimately supporting better crop management and protection strategies.</div></div>","PeriodicalId":52814,"journal":{"name":"Artificial Intelligence in Agriculture","volume":"15 1","pages":"Pages 79-87"},"PeriodicalIF":8.2,"publicationDate":"2025-01-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143097561","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-01-09DOI: 10.1016/j.aiia.2025.01.004
Zhizhong Sun , Jie Yang , Yang Yao , Dong Hu , Yibin Ying , Junxian Guo , Lijuan Xie
Visible/near-infrared (Vis/NIR) spectroscopy technology has been extensively utilized for the determination of soluble solids content (SSC) in fruits. Nonetheless, the spectral distortion resulting from temperature variations in the sample leads to a decrease in detection accuracy. To mitigate the influence of temperature fluctuations on the accuracy of SSC detection in fruits, using watermelon as an example, this study presents a knowledge-guided temperature correction method utilizing one-dimensional convolutional neural networks (1D-CNN). This method consists of two stages: the first stage involves utilizing 1D-CNN models and gradient-weighted class activation mapping (Grad-CAM) method to acquire gradient-weighted features correlating with temperature. The second stage involves mapping these features and integrating them with the original Vis/NIR spectrum, and then train and test the partial least squares (PLS) model. This knowledge-guided method can identify wavelength bands with high temperature correlation in the Vis/NIR spectra, offering valuable guidance for spectral data processing. The performance of the PLS model constructed using the 15 °C spectrum guided by this method is superior to that of the global model, and can reduce the root mean square error of the prediction set (RMSEP) to 0.324°Brix, which is 32.5 % lower than the RMSEP of the global model (0.480°Brix). The method proposed in this study has superior temperature correction effects than slope and bias correction, piecewise direct standardization, and external parameter orthogonalization correction methods. The results indicate that the knowledge-guided temperature correction method based on deep learning can significantly enhance the detection accuracy of SSC in watermelon, providing valuable reference for the development of PLS calibration methods.
可见/近红外（Vis/NIR）光谱技术已广泛应用于水果中可溶性固形物含量的测定。尽管如此，样品中温度变化引起的光谱失真导致检测精度降低。为了减轻温度波动对水果中SSC检测精度的影响，本研究以西瓜为例，提出了一种基于一维卷积神经网络（1D-CNN）的知识引导温度校正方法。该方法分为两个阶段：第一阶段利用1D-CNN模型和梯度加权类激活映射（gradient-weighted class activation mapping, Grad-CAM）方法获取与温度相关的梯度加权特征。第二阶段涉及映射这些特征并将它们与原始的Vis/NIR光谱进行整合，然后训练和测试偏最小二乘（PLS）模型。这种知识引导方法可以识别出可见光/近红外光谱中具有高温相关的波段，为光谱数据处理提供了有价值的指导。利用该方法指导的15°C光谱构建的PLS模型的性能优于全局模型，预测集的均方根误差（RMSEP）降至0.324°Brix，比全局模型的RMSEP（0.480°Brix）低32.5%。该方法的温度校正效果优于斜率和偏置校正、分段直接标准化和外部参数正交化校正方法。结果表明，基于深度学习的知识引导温度校正方法可以显著提高西瓜中SSC的检测精度，为PLS校正方法的开发提供了有价值的参考。
{"title":"Knowledge-guided temperature correction method for soluble solids content detection of watermelon based on Vis/NIR spectroscopy","authors":"Zhizhong Sun , Jie Yang , Yang Yao , Dong Hu , Yibin Ying , Junxian Guo , Lijuan Xie","doi":"10.1016/j.aiia.2025.01.004","DOIUrl":"10.1016/j.aiia.2025.01.004","url":null,"abstract":"<div><div>Visible/near-infrared (Vis/NIR) spectroscopy technology has been extensively utilized for the determination of soluble solids content (SSC) in fruits. Nonetheless, the spectral distortion resulting from temperature variations in the sample leads to a decrease in detection accuracy. To mitigate the influence of temperature fluctuations on the accuracy of SSC detection in fruits, using watermelon as an example, this study presents a knowledge-guided temperature correction method utilizing one-dimensional convolutional neural networks (1D-CNN). This method consists of two stages: the first stage involves utilizing 1D-CNN models and gradient-weighted class activation mapping (Grad-CAM) method to acquire gradient-weighted features correlating with temperature. The second stage involves mapping these features and integrating them with the original Vis/NIR spectrum, and then train and test the partial least squares (PLS) model. This knowledge-guided method can identify wavelength bands with high temperature correlation in the Vis/NIR spectra, offering valuable guidance for spectral data processing. The performance of the PLS model constructed using the 15 °C spectrum guided by this method is superior to that of the global model, and can reduce the root mean square error of the prediction set (RMSEP) to 0.324°Brix, which is 32.5 % lower than the RMSEP of the global model (0.480°Brix). The method proposed in this study has superior temperature correction effects than slope and bias correction, piecewise direct standardization, and external parameter orthogonalization correction methods. The results indicate that the knowledge-guided temperature correction method based on deep learning can significantly enhance the detection accuracy of SSC in watermelon, providing valuable reference for the development of PLS calibration methods.</div></div>","PeriodicalId":52814,"journal":{"name":"Artificial Intelligence in Agriculture","volume":"15 1","pages":"Pages 88-97"},"PeriodicalIF":8.2,"publicationDate":"2025-01-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143097560","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号