This letter presents an efficient and simple image segmentation method for semantic object spatial segmentation. First, the image is filtered using contour-preserving filters. Then it is quasi-flat labeled. The small ...This letter presents an efficient and simple image segmentation method for semantic object spatial segmentation. First, the image is filtered using contour-preserving filters. Then it is quasi-flat labeled. The small regions near the contour are classified as uncertain regions and are eliminated by region growing and merging. Further region merging is used to reduce the region number. The simulation results show its efficiency and simplicity. It can preserve the semantic object shape while emphasize on the perceptual complex part of the object. So it conforms to the human visual perception very well.展开更多
To solve the problems of shaving and reusing information in the information system, a rules-based ontology constructing approach from object-relational databases is proposed. A 3-tuple ontology constructing model is p...To solve the problems of shaving and reusing information in the information system, a rules-based ontology constructing approach from object-relational databases is proposed. A 3-tuple ontology constructing model is proposed first. Then, four types of ontology constructing rules including class, property, property characteristics, and property restrictions ave formalized according to the model. Experiment results described in Web ontology language prove that our proposed approach is feasible for applying in the semantic objects project of semantic computing laboratory in UC Irvine. Our approach reduces about twenty percent constructing time compared with the ontology construction from relational databases.展开更多
This paper presents a cross-media semantic mining model （CSMM） based on object semantic. This model obtains object-level semantic information in terms of maximum probability principle. Then semantic templates are tr...This paper presents a cross-media semantic mining model （CSMM） based on object semantic. This model obtains object-level semantic information in terms of maximum probability principle. Then semantic templates are trained and constructed with STTS （Semantic Template Training System）, which are taken as the bridge to realize the transition from various low-level media feature to object semantic. Furthermore, we put forward a kind of double layers metadata structure to efficaciously store and manage mined low-level feature and high-level semantic. This model has broad application in lots of domains such as intelligent retrieval engine, medical diagnoses, multimedia design and so on.展开更多
For the issue of low positioning accuracy in dynamic environments with traditional simultaneous localisation and mapping(SLAM),a dynamic point removal strategy combining object detection and optical flow tracking has ...For the issue of low positioning accuracy in dynamic environments with traditional simultaneous localisation and mapping(SLAM),a dynamic point removal strategy combining object detection and optical flow tracking has been proposed.To fully utilise the semantic information,an ellipsoid model of the detected semantic objects was first constructed based on the plane and point cloud constraints,which assists in loop closure detection.Bilateral semantic map matching was achieved through the Kuhn-Munkres(KM)algorithm maximum weight assignment,and the pose transformation between local and global maps was determined by the random sample consensus(RANSAC)algorithm.Finally,a stable semantic SLAM system suitable for dy-namic environments was constructed.The effectiveness of achieving the system's positioning accuracy under dynamic inter-ference and large visual-inertial loop closure was verified by the experiment.展开更多
Weed management plays a crucial role in increasing crop yields.Semantic segmentation,which classifies each pixel in an image captured by a camera into categories such as crops,weeds,and background,is a widely used met...Weed management plays a crucial role in increasing crop yields.Semantic segmentation,which classifies each pixel in an image captured by a camera into categories such as crops,weeds,and background,is a widely used method in this context.However,conventional semantic segmentation methods rely solely on pixel information within the camera's field of view(FOV),hindering their ability to detect weeds outside the visible area.This limitation can lead to incomplete weed removal and inefficient herbicide application.Incorporating information beyond the FOV in crop and weed segmentation is therefore essential for effective herbicide usage.Nevertheless,existing research on crop and weed segmentation has largely overlooked this limitation.To address this issue,we propose the knowledge distillation-based outpainting and semantic segmentation network(KDOSS-Net)for crop and weed images,a novel framework that enhances segmentation accuracy by leveraging information beyond the FOV.KDOSS-Net consists of two parts:the object prediction-guided outpainting and semantic segmentation network(OPOSS-Net),which serves as the teacher model by restoring areas outside the FOV and performing semantic segmentation,and the semantic segmentation without outpainting network(SSWO-Net),which serves as the student model,directly performing segmentation without outpainting.Through knowledge distillation(KD),the student model learns from the teacher's outputs,which results in a lightweight yet highly accurate segmentation network that is suitable for deployment on agricultural robots with limited computing power.Experiments on three public datasets-Rice seedling and weed,CWFID,and BoniRob-yielded mean intersection over union(mIOU)scores of 0.6315,0.7101,and 0.7524,respectively.These results demonstrate that KDOSS-Net achieves higher accuracy than existing state-of-the-art(SOTA)segmentation models while significantly reducing computational overhead.Furthermore,the weed information extracted using our method is automatically linked as input to the open-source large language and vision assistant(LLaVA),enabling the development of a system that recommends optimal herbicide strategies tailored to the detected weed class.展开更多
基金Supported by Guangdong Natural Science Foundation(No.011628)
文摘This letter presents an efficient and simple image segmentation method for semantic object spatial segmentation. First, the image is filtered using contour-preserving filters. Then it is quasi-flat labeled. The small regions near the contour are classified as uncertain regions and are eliminated by region growing and merging. Further region merging is used to reduce the region number. The simulation results show its efficiency and simplicity. It can preserve the semantic object shape while emphasize on the perceptual complex part of the object. So it conforms to the human visual perception very well.
基金supported by the National Natural Science Foundation of China (60471055)the National "863" High Technology Research and Development Program of China (2007AA01Z443)
文摘To solve the problems of shaving and reusing information in the information system, a rules-based ontology constructing approach from object-relational databases is proposed. A 3-tuple ontology constructing model is proposed first. Then, four types of ontology constructing rules including class, property, property characteristics, and property restrictions ave formalized according to the model. Experiment results described in Web ontology language prove that our proposed approach is feasible for applying in the semantic objects project of semantic computing laboratory in UC Irvine. Our approach reduces about twenty percent constructing time compared with the ontology construction from relational databases.
基金Supported by the National Basic Research Program of China 973 Program (2007CB310801)the Specialized Research Fund for the Doctoral Program of Higer Education of China (20070486064)+1 种基金the Natural Science Foundation of Hubei Province (2007ABA038)the Programme of Introducing Talents of Discipline to Universities (B07037)
文摘This paper presents a cross-media semantic mining model （CSMM） based on object semantic. This model obtains object-level semantic information in terms of maximum probability principle. Then semantic templates are trained and constructed with STTS （Semantic Template Training System）, which are taken as the bridge to realize the transition from various low-level media feature to object semantic. Furthermore, we put forward a kind of double layers metadata structure to efficaciously store and manage mined low-level feature and high-level semantic. This model has broad application in lots of domains such as intelligent retrieval engine, medical diagnoses, multimedia design and so on.
基金supported in part by the Natural Science Foundation of Shandong Province(No.ZR2024MF036)the National Key Research and Development Plan of China(No.2020AAA0109000)the National Natural Science Foundation of China(Nos.61973184,61803227,61603214,and 61573213).
文摘For the issue of low positioning accuracy in dynamic environments with traditional simultaneous localisation and mapping(SLAM),a dynamic point removal strategy combining object detection and optical flow tracking has been proposed.To fully utilise the semantic information,an ellipsoid model of the detected semantic objects was first constructed based on the plane and point cloud constraints,which assists in loop closure detection.Bilateral semantic map matching was achieved through the Kuhn-Munkres(KM)algorithm maximum weight assignment,and the pose transformation between local and global maps was determined by the random sample consensus(RANSAC)algorithm.Finally,a stable semantic SLAM system suitable for dy-namic environments was constructed.The effectiveness of achieving the system's positioning accuracy under dynamic inter-ference and large visual-inertial loop closure was verified by the experiment.
基金This work was supported in part by the Ministry of Science and ICT(MSIT),Korea,through the Information Technology Research Center(ITRC)Support Program under Grant IITP-2025-RS-2020-II201789in part by the Artificial Intelligence Convergence Innovation Human Resources Development Supervised by the Institute of Information&Communications Technology Planning&Evaluation(IITP)under Grant IITP-2025-RS-2023-00254592.
文摘Weed management plays a crucial role in increasing crop yields.Semantic segmentation,which classifies each pixel in an image captured by a camera into categories such as crops,weeds,and background,is a widely used method in this context.However,conventional semantic segmentation methods rely solely on pixel information within the camera's field of view(FOV),hindering their ability to detect weeds outside the visible area.This limitation can lead to incomplete weed removal and inefficient herbicide application.Incorporating information beyond the FOV in crop and weed segmentation is therefore essential for effective herbicide usage.Nevertheless,existing research on crop and weed segmentation has largely overlooked this limitation.To address this issue,we propose the knowledge distillation-based outpainting and semantic segmentation network(KDOSS-Net)for crop and weed images,a novel framework that enhances segmentation accuracy by leveraging information beyond the FOV.KDOSS-Net consists of two parts:the object prediction-guided outpainting and semantic segmentation network(OPOSS-Net),which serves as the teacher model by restoring areas outside the FOV and performing semantic segmentation,and the semantic segmentation without outpainting network(SSWO-Net),which serves as the student model,directly performing segmentation without outpainting.Through knowledge distillation(KD),the student model learns from the teacher's outputs,which results in a lightweight yet highly accurate segmentation network that is suitable for deployment on agricultural robots with limited computing power.Experiments on three public datasets-Rice seedling and weed,CWFID,and BoniRob-yielded mean intersection over union(mIOU)scores of 0.6315,0.7101,and 0.7524,respectively.These results demonstrate that KDOSS-Net achieves higher accuracy than existing state-of-the-art(SOTA)segmentation models while significantly reducing computational overhead.Furthermore,the weed information extracted using our method is automatically linked as input to the open-source large language and vision assistant(LLaVA),enabling the development of a system that recommends optimal herbicide strategies tailored to the detected weed class.
