检索结果-内蒙古大学图书馆

Comparing YOLOv8 and Mask R-CNN for instance segmentation in complex orchard environments

artificial intelligence in Agriculture 2024年第3期13卷 84-99页

作者： Ranjan Sapkota Dawood Ahmed Manoj Karkee Center for Precision&Automated Agricultural Systems Washington State University24106 N Bunn RdProsser99350 WashingtonUSA

Instance segmentation,an important image processing operation for automation in agriculture,is used to precisely delineate individual objects of interestwithin images,which provides foundational information for various automated or robotic tasks such as selective harvesting and precision *** study compares the one-stage YOLOv8 and the two-stage Mask R-CNN machine learning models for instance segmentation under varying orchard conditions across two *** 1,collected in dormant season,includes images of dormant apple trees,which were used to train multi-object segmentation models delineating tree branches and *** 2,collected in the early growing season,includes images of apple tree canopies with green foliage and immature(green)apples(also called fruitlet),which were used to train single-object segmentation models delineating only immature green *** results showed that YOLOv8 performed better than Mask R-CNN,achieving good precision and near-perfect recall across both datasets at a confidence threshold of ***,for Dataset 1,YOLOv8 achieved a precision of 0.90 and a recall of 0.95 for all *** comparison,Mask R-CNN demonstrated a precision of 0.81 and a recall of 0.81 for the *** Dataset 2,YOLOv8 achieved a precision of 0.93 and a recall of *** R-CNN,in this single-class scenario,achieved a precision of 0.85 and a recall of ***,the inference times for YOLOv8 were 10.9 ms for multi-class segmentation(Dataset 1)and 7.8 ms for single-class segmentation(Dataset 2),compared to 15.6 ms and 12.8 ms achieved by Mask R-CNN's,*** findings showYOLOv8's superior accuracy and efficiency in machine learning applications compared to two-stage models,specifically Mask-R-CNN,which suggests its suitability in developing smart and automated orchard operations,particularly when real-time applications are necessary in such cases as robotic harvesting and robotic immature green fruit thin

关键词： YOLOv8 Mask R-CNN Deep learning machine learning Automation robotics artificial intelligence machine vision

来源：评论

学校读者我要写书评

暂无评论

Recent Progress in High-performance Flexible Pressure Sensors: Materials, Design, Fabrication, and applications

引用

ADVANCED MATERIALS TECHNOLOGIES 2025年

作者： xu, Dandan Zhou, Peilin Wu, Jiachen xiang, Guangrui Zhong, Ya Guo, Hongji Zhang, xiuli xia, Yueqing Yu, Haibo Liu, Lianqing Henan Agr Univ Coll Mech & Elect Engn Zhengzhou 450002 Peoples R China Chinese Acad Sci Shenyang Inst Automat State Key Lab Robot Shenyang 110016 Peoples R China

Flexible pressure sensors hold significant potential for applications in health monitoring, human-machine interaction, electronic skin, and artificial intelligence due to their high sensitivity, flexibility, lightweight, and ease of signal acquisition. In recent years, extensive research into sensor materials, structures, and manufacturing technologies has led to the development of various high-performance flexible pressure sensors. Currently, optimizing sensing performance involves selecting appropriate functional materials, designing deformable structures, and employing high-precision manufacturing techniques. This paper reviews recent advancements in flexible pressure sensors, focusing on sensing mechanisms, functional materials, microstructure design, manufacturing technologies, and application fields. First, the sensing mechanisms of pressure sensors operating in different modes are introduced, and several widely used functional materials are discussed. Particular attention is given to the role of geometric microstructure design in enhancing sensing performance. Next, the influence of various manufacturing technologies on sensing performance is analyzed and summarized. In addition, emerging applications of flexible pressure sensors in health monitoring, human-machine interaction, electronic skin, and artificial intelligence are presented. Finally, the paper concludes by highlighting the development prospects and major challenges in achieving high-performance flexible pressure sensors.

关键词： applications flexible pressure sensor functional materials manufacturing technologies structures

来源：评论

学校读者我要写书评

暂无评论

Computer simulation and hands-on labs: A case study of teaching robotics and AI

引用

INTERNATIONAL JOURNAL OF MECHANICAL ENGINEERING EDUCATION 2025年第2期53卷 451-469页

作者： Ubando, Luis Alberto Munoz Amigud, Alexander Sirazitdinova, Ekaterina Tec Monterrey Sch Engn & Sci Monterrey Mexico TEC Monterrey Inst Future Educ Monterrey Mexico NVIDIA Corp Munich Germany

When teaching robotics, instructors face the challenge of finding an effective approach to bridge theoretical concepts and practical applications. Both computer simulations and hands-on laboratory experiments provide learners with opportunities for active, immersive, and experiential learning. As students progress from introductory to advanced topics and from theory to practice, their performance is contingent upon earlier knowledge and may increase, remain unchanged, or decrease. The question that arises is whether computer simulation can serve as a viable foundation for fostering an understanding of theory that enables the subsequent grasp of advanced practical concepts in robotics. Put another way, when students are introduced to the field of robotics through computer simulation, how will they perform when presented with advanced hands-on tasks involving the construction of physical robots to solve problems in physical space? To answer this question, we examined undergraduate student performance (n = 107) across two robotics courses-an introductory course using computer simulation (Robot Operating System, Rviz, and GAZEBO) and an advanced course using physical hardware (Puzzlebot), leveraging the hardware's capability for AI tasks such as machine vision (Nvidia Jetson Nano development kit). Our findings suggest that student performance increased as they progressed from using computer simulation to engaging with hardware in the physical environment, further suggesting that teaching with computer simulations provides an adequate foundation to learn and complete more advanced tasks.

关键词： robotics science education active learning computer simulation artificial intelligence

来源：评论

学校读者我要写书评

暂无评论

CLIP Based Semantic Information Extraction and Target Alignment for Domain Generalization 17th

CLIP Based Semantic Information Extraction and Target Alignm...

引用

17th International Conference on Intelligent robotics and applications

作者： Li, Yuxi Zuo, Weiliang Chen, Zikai xin, Jingmin Zheng, Nanning Xi An Jiao Tong Univ Natl Key Lab Human Machine Hybrid Augmented Intel Natl Engn Res Ctr Visual Informat & Applicat Xian Peoples R China Xi An Jiao Tong Univ Inst Artificial Intelligence & Robot Xian Peoples R China

ISBN: (纸本)9789819607853;9789819607860

machine learning models suffer from serious performance declination when facing out-of-distribution datasets. In recent years, numerous researches on domain generalization (DG) have been made to address that issue and improve the model's generalization. With the development of multimodal models, there are an increasing number of works considering utilizing large vision-language models to achieve DG. In this paper, we proposed Semantic Information Extraction and Target Alignment (SIETA) to take alignment both in the training and testing phases. We choose pre-trained CLIP as the teacher to guide our model to learn the ability to extract semantic information by minimizing the distance between the representations of CLIP's text encoder and those of our encoder during training and use knowledge distillation to transfer the abundant prior knowledge of CLIP into our model. In the inference phase, we leverage Test-Time Adaptation (TTA) to slightly align our model with the target domain to further enhance the model's generalization. We conducted experiments on four DG benchmark datasets, and the results show that our method significantly improves the model's generalization with a smaller size than CLIP and is versatile to combine with other DG methods.

关键词： Domain generalization Semantic information extraction CLIP

来源：评论

学校读者我要写书评

暂无评论

International Conference on Control and Automation, Electronics, robotics, Internet of Things, and artificial intelligence, CERIA 2024

International Conference on Control and Automation, Electron...

引用

2024 International Conference on Control and Automation, Electronics, robotics, Internet of Things, and artificial intelligence, CERIA 2024

ISBN: (纸本)9798331511166

The proceedings contain 53 papers. The topics discussed include: understanding Internet usage patterns and behaviors among college students: a machine learning approach;applications of wireless sensor networks in biomedical research and healthcare;development of obstacle navigation in robots using the linear regression method;estimation of rock compressibility for Indonesian limestones by artificial neural network;leaf based grape variety identification using computer vision;implementation of convolutional neural network in medical robot facial recognition system;recent applications using explainable artificial intelligence for data analytics;and development of an autonomous navigation system using google maps API for a package delivery device.

关键词：

来源：评论

学校读者我要写书评

暂无评论

ChatGPT: vision and Challenges

引用

arxiv 2023年第1期3卷 262-271页

作者： Gill, Sukhpal Singh Kaur, Rupinder School of Electronic Engineering and Computer Science Queen Mary University of London United Kingdom Department of Science Kings Education London United Kingdom

artificial intelligence (AI) and machine learning have changed the nature of scientific inquiry in recent years. Of these, the development of virtual assistants has accelerated greatly in the past few years, with ChatGPT becoming a prominent AI language model. In this study, we examine the foundations, vision, research challenges of ChatGPT. This article investigates into the background and development of the technology behind it, as well as its popular applications. Moreover, we discuss the advantages of bringing everything together through ChatGPT and Internet of Things (IoT). Further, we speculate on the future of ChatGPT by considering various possibilities for study and development, such as energy-efficiency, cybersecurity, enhancing its applicability to additional technologies (robotics and Computer vision), strengthening human-AI communications, and bridging the technological gap. Finally, we discuss the important ethics and current trends of ChatGPT. © 2023, CC0.

关键词： machine learning

来源：评论

学校读者我要写书评

暂无评论

machine vision applications in robotics and Automation 9

Machine Vision Applications in Robotics and Automation

引用

9th International Conference on Science, Technology, Engineering and Mathematics, ICONSTEM 2024

作者： Bharathi, G.P. Kavitha, P. Abhang, Laxman Baburao Rathinasabapathi, G. Sharma, Vipin Sathish, T. Sri Sai Ram Institute of Technology Department of Computer And Communication Engineering West Tambaram Tamil Nadu Chennai600045 India R.M.K Engineering College Department of Electrical And Electronics Engineering Tamil Nadu Kavaraipettai601206 India Pravara Rural Engineering College Department of Automation And Robotics Maharashtra Loni413736 India Panimalar Engineering College Department of Mechanical Engineering Tamil Nadu Chennai600123 India Medi-Caps University Department of Mechanical Engineering Madhya Pradesh Indore453331 India Saveetha School of Engineering Simats Department of Mechanical Engineering Tamil Nadu Chennai602105 India

ISBN: (纸本)9798350365092

The extensive incorporation of machine vision into the fields of robotics and automation in a variety of different ways. The various uses of machine vision and the revolutionary impact it has on the capabilities of robotic systems. This investigation is fueled by improvements in imaging technology and artificial intelligence. In the abstract, major features such as object detection, navigation, and human-robot interaction are discussed in depth. The abstract also demonstrates how machine vision algorithms contribute to enhanced precision, efficiency, and adaptability in automation processes. The purpose of thisis to shed light on the role that machine vision plays in revolutionising industrial automation, collaborative robotics, and autonomous systems by conducting an examination of real-world implementations and case studies. This all-encompassing investigation not only highlights the current status of machine vision applications but also forecasts the future trajectory of this technology, thereby establishing it as a foundational component for the development of robotics and automation in a variety of fields. © 2024 IEEE.

关键词： Navigation systems

来源：评论

学校读者我要写书评

暂无评论

Transformer-Based Sensor Fusion for Autonomous Vehicles: A Comprehensive Review

引用

IEEE ACCESS 2025年 13卷 41822-41838页

作者： Abdulmaksoud, Ahmed Ahmed, Ryan McMaster Univ Ctr Mechatron & Hybrid Technol CMHT Dept Mech Engn Hamilton ON L8S 4L8 Canada

Sensor fusion is vital for many critical applications, including robotics, autonomous driving, aerospace, and beyond. Integrating data streams from different sensors enables us to overcome the intrinsic limitations of each sensor, providing more reliable measurements and reducing uncertainty. Moreover, deep learning-based sensor fusion unlocked the possibility of multimodal learning, which utilizes different sensor modalities to boost object detection. Yet, adverse weather conditions remain a significant challenge to the reliability of sensor fusion. However, introducing the Transformer deep learning model in sensor fusion presents a promising avenue for advancing its sensing capabilities, potentially overcoming that challenge. Transformer models proved powerful in modeling vision, language, and numerous other domains. However, these models suffer from high latency and heavy computation requirements. This paper aims to provide: 1) an extensive overview of sensor fusion and transformer models;2) an in-depth survey of the state-of-the-art (SoTA) methods for Transformer-based sensor fusion, focusing on camera-LiDAR and camera-radar methods;and 3) a quantitative analysis of the SoTA methods, uncovering research gaps and stimulating future work.

关键词： Sensor fusion Transformers Laser radar Autonomous vehicles Cameras Computational modeling Reviews Three-dimensional displays Object detection Location awareness Autonomous driving artificial intelligence (AI) computer vision deep learning machine learning sensor fusion transformers

来源：评论

学校读者我要写书评

暂无评论

A fast, lightweight deep learning vision pipeline for autonomous UAV landing support with added robustness

引用

ENGINEERING applications OF artificial intelligence 2024年 131卷

作者： Pieczynski, Dominik Ptak, Bartosz Kraft, Marek Piechocki, Mateusz Aszkowski, Przemyslaw Poznan Univ Tech Inst Robot & Machine Intelligence Piotrowo 3A PL-60965 Poznan Poland

Despite massive development in aerial robotics, precise and autonomous landing in various conditions is still challenging. This process is affected by many factors, such as terrain shape, weather conditions, and the presence of obstacles. This paper describes a deep learning-accelerated image processing pipeline for accurate detection and relative pose estimation of the UAV with respect to the landing pad. Moreover, the system provides increased safety and robustness by implementing human presence detection and error estimation for both landing target detection and pose computation. Human presence and landing pad location are performed by estimating the presence probability via segmentation. This is followed by the landing pad keypoints' location regression algorithm, which, in addition to coordinates, provides the uncertainty of presence for each defined landing pad landmark. To perform the aforementioned tasks, a set of lightweight neural network models was selected and evaluated. The resulting measurements of the system's performance and accuracy are presented for each component individually and for the whole processing pipeline. The measurements are performed using onboard embedded UAV hardware and confirm that the method can provide accurate, low-latency feedback information for safe landing support.

关键词： Unmanned aerial vehicle Landing support Image processing Deep learning On-board processing

来源：评论

学校读者我要写书评

暂无评论

Embracing robotics and Intelligent machine Systems for Smart Agricultural applications [From the Guest Editors]

引用

IEEE robotics & AUTOMATION MAGAZINE 2023年第4期30卷 8页

作者： van Henten, Eldert Montenegro, Carlos Popovic, Marija Vougioukas, Stavros Daniel, Alfred Han, Guangjie

High tech to feed the world! This slogan was launched some 10 years ago in The Netherlands to direct research on high tech for agricultural production. It responded to an urgent concern about the needs of a growing world population for food, feed, fibers, and basic chemical components. By the way, make no mistake, agriculture does not only provide food! Anyway, the slogan represented the vibrant optimism about the potential of technology in support of this quest. And for good reasons! Advances in robotics and artificial intelligence (AI) in the recent past allowed us to go beyond far-fetched visions and speculations and put these technologies to the test on the fields, in the greenhouses, in the orchards, and in livestock barns. Definitely, this trend is not limited to The Netherlands. When monitoring academic and professional literature, the rapidly growing number of publications in the past 10 years addressing robotics and AI in the agricultural context is clear proof of a worldwide trend.

关键词： Special issues and sections Smart agriculture Crop yield Farming Food products artificial intelligence

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：