Lane detection is a fundamental aspect of most current advanced driver assistance systems(ADASs). A large number of existing results focus on the study of vision-based lane detection methods due to the extensive knowl...Lane detection is a fundamental aspect of most current advanced driver assistance systems(ADASs). A large number of existing results focus on the study of vision-based lane detection methods due to the extensive knowledge background and the low-cost of camera devices. In this paper, previous visionbased lane detection studies are reviewed in terms of three aspects, which are lane detection algorithms, integration, and evaluation methods. Next, considering the inevitable limitations that exist in the camera-based lane detection system, the system integration methodologies for constructing more robust detection systems are reviewed and analyzed. The integration methods are further divided into three levels, namely, algorithm, system,and sensor. Algorithm level combines different lane detection algorithms while system level integrates other object detection systems to comprehensively detect lane positions. Sensor level uses multi-modal sensors to build a robust lane recognition system. In view of the complexity of evaluating the detection system, and the lack of common evaluation procedure and uniform metrics in past studies, the existing evaluation methods and metrics are analyzed and classified to propose a better evaluation of the lane detection system. Next, a comparison of representative studies is performed. Finally, a discussion on the limitations of current lane detection systems and the future developing trends toward an Artificial Society, Computational experiment-based parallel lane detection framework is proposed.展开更多
In the area of computer vision, deep learning has produced a variety of state-of-the-art models that rely on massive labeled data. However, collecting and annotating images from the real world is too demanding in term...In the area of computer vision, deep learning has produced a variety of state-of-the-art models that rely on massive labeled data. However, collecting and annotating images from the real world is too demanding in terms of labor and money investments, and is usually inflexible to build datasets with specific characteristics, such as small area of objects and high occlusion level. Under the framework of Parallel Vision, this paper presents a purposeful way to design artificial scenes and automatically generate virtual images with precise annotations.A virtual dataset named Parallel Eye is built, which can be used for several computer vision tasks. Then, by training the DPM(Deformable parts model) and Faster R-CNN detectors, we prove that the performance of models can be significantly improved by combining Parallel Eye with publicly available real-world datasets during the training phase. In addition, we investigate the potential of testing the trained models from a specific aspect using intentionally designed virtual datasets, in order to discover the flaws of trained models. From the experimental results, we conclude that our virtual dataset is viable to train and test the object detectors.展开更多
目的视觉感知技术是智能车系统中的一项关键技术,但是在复杂挑战下如何有效提高视觉性能已经成为智能驾驶领域的重要研究内容。本文将人工社会(artificial societies)、计算实验(computational experiments)和平行执行(parallel executi...目的视觉感知技术是智能车系统中的一项关键技术,但是在复杂挑战下如何有效提高视觉性能已经成为智能驾驶领域的重要研究内容。本文将人工社会(artificial societies)、计算实验(computational experiments)和平行执行(parallel execution)构成的ACP方法引入智能驾驶的视觉感知领域,提出了面向智能驾驶的平行视觉感知,解决了视觉模型合理训练和评估问题,有助于智能驾驶进一步走向实际应用。方法平行视觉感知通过人工子系统组合来模拟实际驾驶场景,构建人工驾驶场景使之成为智能车视觉感知的"计算实验室";借助计算实验两种操作模式完成视觉模型训练与评估;最后采用平行执行动态优化视觉模型,保障智能驾驶对复杂挑战的感知与理解长期有效。结果实验表明,目标检测的训练阶段虚实混合数据最高精度可达60.9%,比单纯用KPC(包括:KITTI(Karlsruhe Institute of Technology and Toyota Technological Institute),PASCAL VOC(pattern analysis,statistical modelling and computational learning visual object classes)和MS COCO(Microsoft common objects in context))数据和虚拟数据分别高出17.9%和5.3%;在评估阶段相较于基准数据,常规任务(-30°且垂直移动)平均精度下降11.3%,环境任务(雾天)平均精度下降21.0%,困难任务(所有挑战)平均精度下降33.7%。结论本文为智能驾驶设计和实施了在实际驾驶场景难以甚至无法进行的视觉计算实验,对复杂视觉挑战进行分析和评估,具备加强智能车在行驶过程中感知和理解周围场景的意义。展开更多
针对通用目标检测方法YOLO(you only look once)直接应用到人脸检测中存在召回率不够高、定位不够准确的问题,提出一种由密集到稀疏的多尺度并行的网络结构。通过不同尺度的网络检测不同尺寸的人脸,解决召回率不够高的问题,通过平均多...针对通用目标检测方法YOLO(you only look once)直接应用到人脸检测中存在召回率不够高、定位不够准确的问题,提出一种由密集到稀疏的多尺度并行的网络结构。通过不同尺度的网络检测不同尺寸的人脸,解决召回率不够高的问题,通过平均多尺度网络的检测结果解决定位不够准确的问题。引入中心损失函数,减小类内距离,进一步提高分类准确率。实验结果表明,在不同的数据集上,该方法的召回率及定位准确性相对于YOLO有所提高,检测精度接近主流方法的同时检测速度具有明显优势。展开更多
In this paper,a combined robust fault detection and isolation scheme is studied for satellite system subject to actuator faults,external disturbances,and parametric uncertainties.The proposed methodology incorporates ...In this paper,a combined robust fault detection and isolation scheme is studied for satellite system subject to actuator faults,external disturbances,and parametric uncertainties.The proposed methodology incorporates a residual generation module,including a bank of filters,into an intelligent residual evaluation module.First,residual filters are designed based on an improved nonlinear differential algebraic approach so that they are not affected by external disturbances.The residual evaluation module is developed based on the suggested series and parallel forms.Further,a new ensemble classification scheme defined as blended learning integrates heterogeneous classifiers to enhance the performance.A wide range of simulations is carried out in a high-fidelity satellite simulator subject to the constant and time-varying actuator faults in the presence of disturbances,manoeuvres,uncertainties,and noises.The obtained results demonstrate the effectiveness of the proposed robust fault detection and isolation method compared to the traditional nonlinear differential algebraic approach.展开更多
A novel image encryption scheme based on parallel compressive sensing and edge detection embedding technology is proposed to improve visual security. Firstly, the plain image is sparsely represented using the discrete...A novel image encryption scheme based on parallel compressive sensing and edge detection embedding technology is proposed to improve visual security. Firstly, the plain image is sparsely represented using the discrete wavelet transform.Then, the coefficient matrix is scrambled and compressed to obtain a size-reduced image using the Fisher–Yates shuffle and parallel compressive sensing. Subsequently, to increase the security of the proposed algorithm, the compressed image is re-encrypted through permutation and diffusion to obtain a noise-like secret image. Finally, an adaptive embedding method based on edge detection for different carrier images is proposed to generate a visually meaningful cipher image. To improve the plaintext sensitivity of the algorithm, the counter mode is combined with the hash function to generate keys for chaotic systems. Additionally, an effective permutation method is designed to scramble the pixels of the compressed image in the re-encryption stage. The simulation results and analyses demonstrate that the proposed algorithm performs well in terms of visual security and decryption quality.展开更多
Considering the current environmental pollution caused by chemical topping,plant damage caused by mechanical topping,and the high cost of manual topping,a laser-based cotton-tip pruning robot for field cotton was desi...Considering the current environmental pollution caused by chemical topping,plant damage caused by mechanical topping,and the high cost of manual topping,a laser-based cotton-tip pruning robot for field cotton was designed in this study.The main advantages of this robot include its safety,light weight,low cost,and environmental friendliness.First,the structural design,measurement,and corresponding control system design of the robot were realized.Subsequently,a precise laser control method based on Yolov5 cotton top identification and the inverse kinematic solution of parallel robot rapid positioning were examined.Subsequently,the optimal laser irradiation wavelength and duration were determined.Finally,a laser-topping experiment was conducted,and the overall accuracy and recall rates for cotton identification were 98.3%and 99.3%,respectively.The AP and mAP at the threshold value of 0.5 reached 99.3%and 78.8%,respectively.The maximum positioning error of the jacking system is 2.6 mm,and the repeated positioning error is within±1.2 cm,which meets the accuracy requirements of laser jacking.Blue-purple laser irradiation at 15 W and 405 nm for 22 s was the best topping scheme.Comparing the effects of cotton topping with and without manual topping,it can be seen that laser topping significantly improved the yield and quality.Combined with the photothermal response of the cotton topping,the feasibility of laser topping was verified both theoretically and experimentally.The laser-topping scheme proposed in this study exhibits high efficiency,environmental protection,and safety,as well as good application prospects.展开更多
文摘Lane detection is a fundamental aspect of most current advanced driver assistance systems(ADASs). A large number of existing results focus on the study of vision-based lane detection methods due to the extensive knowledge background and the low-cost of camera devices. In this paper, previous visionbased lane detection studies are reviewed in terms of three aspects, which are lane detection algorithms, integration, and evaluation methods. Next, considering the inevitable limitations that exist in the camera-based lane detection system, the system integration methodologies for constructing more robust detection systems are reviewed and analyzed. The integration methods are further divided into three levels, namely, algorithm, system,and sensor. Algorithm level combines different lane detection algorithms while system level integrates other object detection systems to comprehensively detect lane positions. Sensor level uses multi-modal sensors to build a robust lane recognition system. In view of the complexity of evaluating the detection system, and the lack of common evaluation procedure and uniform metrics in past studies, the existing evaluation methods and metrics are analyzed and classified to propose a better evaluation of the lane detection system. Next, a comparison of representative studies is performed. Finally, a discussion on the limitations of current lane detection systems and the future developing trends toward an Artificial Society, Computational experiment-based parallel lane detection framework is proposed.
基金supported by the National Natural Science Foundation of China(61533019,71232006)
文摘In the area of computer vision, deep learning has produced a variety of state-of-the-art models that rely on massive labeled data. However, collecting and annotating images from the real world is too demanding in terms of labor and money investments, and is usually inflexible to build datasets with specific characteristics, such as small area of objects and high occlusion level. Under the framework of Parallel Vision, this paper presents a purposeful way to design artificial scenes and automatically generate virtual images with precise annotations.A virtual dataset named Parallel Eye is built, which can be used for several computer vision tasks. Then, by training the DPM(Deformable parts model) and Faster R-CNN detectors, we prove that the performance of models can be significantly improved by combining Parallel Eye with publicly available real-world datasets during the training phase. In addition, we investigate the potential of testing the trained models from a specific aspect using intentionally designed virtual datasets, in order to discover the flaws of trained models. From the experimental results, we conclude that our virtual dataset is viable to train and test the object detectors.
文摘目的视觉感知技术是智能车系统中的一项关键技术,但是在复杂挑战下如何有效提高视觉性能已经成为智能驾驶领域的重要研究内容。本文将人工社会(artificial societies)、计算实验(computational experiments)和平行执行(parallel execution)构成的ACP方法引入智能驾驶的视觉感知领域,提出了面向智能驾驶的平行视觉感知,解决了视觉模型合理训练和评估问题,有助于智能驾驶进一步走向实际应用。方法平行视觉感知通过人工子系统组合来模拟实际驾驶场景,构建人工驾驶场景使之成为智能车视觉感知的"计算实验室";借助计算实验两种操作模式完成视觉模型训练与评估;最后采用平行执行动态优化视觉模型,保障智能驾驶对复杂挑战的感知与理解长期有效。结果实验表明,目标检测的训练阶段虚实混合数据最高精度可达60.9%,比单纯用KPC(包括:KITTI(Karlsruhe Institute of Technology and Toyota Technological Institute),PASCAL VOC(pattern analysis,statistical modelling and computational learning visual object classes)和MS COCO(Microsoft common objects in context))数据和虚拟数据分别高出17.9%和5.3%;在评估阶段相较于基准数据,常规任务(-30°且垂直移动)平均精度下降11.3%,环境任务(雾天)平均精度下降21.0%,困难任务(所有挑战)平均精度下降33.7%。结论本文为智能驾驶设计和实施了在实际驾驶场景难以甚至无法进行的视觉计算实验,对复杂视觉挑战进行分析和评估,具备加强智能车在行驶过程中感知和理解周围场景的意义。
文摘针对通用目标检测方法YOLO(you only look once)直接应用到人脸检测中存在召回率不够高、定位不够准确的问题,提出一种由密集到稀疏的多尺度并行的网络结构。通过不同尺度的网络检测不同尺寸的人脸,解决召回率不够高的问题,通过平均多尺度网络的检测结果解决定位不够准确的问题。引入中心损失函数,减小类内距离,进一步提高分类准确率。实验结果表明,在不同的数据集上,该方法的召回率及定位准确性相对于YOLO有所提高,检测精度接近主流方法的同时检测速度具有明显优势。
文摘In this paper,a combined robust fault detection and isolation scheme is studied for satellite system subject to actuator faults,external disturbances,and parametric uncertainties.The proposed methodology incorporates a residual generation module,including a bank of filters,into an intelligent residual evaluation module.First,residual filters are designed based on an improved nonlinear differential algebraic approach so that they are not affected by external disturbances.The residual evaluation module is developed based on the suggested series and parallel forms.Further,a new ensemble classification scheme defined as blended learning integrates heterogeneous classifiers to enhance the performance.A wide range of simulations is carried out in a high-fidelity satellite simulator subject to the constant and time-varying actuator faults in the presence of disturbances,manoeuvres,uncertainties,and noises.The obtained results demonstrate the effectiveness of the proposed robust fault detection and isolation method compared to the traditional nonlinear differential algebraic approach.
基金supported by the Key Area R&D Program of Guangdong Province (Grant No.2022B0701180001)the National Natural Science Foundation of China (Grant No.61801127)+1 种基金the Science Technology Planning Project of Guangdong Province,China (Grant Nos.2019B010140002 and 2020B111110002)the Guangdong-Hong Kong-Macao Joint Innovation Field Project (Grant No.2021A0505080006)。
文摘A novel image encryption scheme based on parallel compressive sensing and edge detection embedding technology is proposed to improve visual security. Firstly, the plain image is sparsely represented using the discrete wavelet transform.Then, the coefficient matrix is scrambled and compressed to obtain a size-reduced image using the Fisher–Yates shuffle and parallel compressive sensing. Subsequently, to increase the security of the proposed algorithm, the compressed image is re-encrypted through permutation and diffusion to obtain a noise-like secret image. Finally, an adaptive embedding method based on edge detection for different carrier images is proposed to generate a visually meaningful cipher image. To improve the plaintext sensitivity of the algorithm, the counter mode is combined with the hash function to generate keys for chaotic systems. Additionally, an effective permutation method is designed to scramble the pixels of the compressed image in the re-encryption stage. The simulation results and analyses demonstrate that the proposed algorithm performs well in terms of visual security and decryption quality.
基金supported in part by the Modern Agriculture Project of Jiangsu Province Science and Technology Plan Special Fund Project(Grant No.BE2022363)the Fundamental Research Funds for the Central Univerisities(Grant No.KYCXJC2024003)the Modern Agricultural Machinery Equipment and Technology Demonstration and Promotion Project of Jiangsu Province(Grant No.NJ2022-03).
文摘Considering the current environmental pollution caused by chemical topping,plant damage caused by mechanical topping,and the high cost of manual topping,a laser-based cotton-tip pruning robot for field cotton was designed in this study.The main advantages of this robot include its safety,light weight,low cost,and environmental friendliness.First,the structural design,measurement,and corresponding control system design of the robot were realized.Subsequently,a precise laser control method based on Yolov5 cotton top identification and the inverse kinematic solution of parallel robot rapid positioning were examined.Subsequently,the optimal laser irradiation wavelength and duration were determined.Finally,a laser-topping experiment was conducted,and the overall accuracy and recall rates for cotton identification were 98.3%and 99.3%,respectively.The AP and mAP at the threshold value of 0.5 reached 99.3%and 78.8%,respectively.The maximum positioning error of the jacking system is 2.6 mm,and the repeated positioning error is within±1.2 cm,which meets the accuracy requirements of laser jacking.Blue-purple laser irradiation at 15 W and 405 nm for 22 s was the best topping scheme.Comparing the effects of cotton topping with and without manual topping,it can be seen that laser topping significantly improved the yield and quality.Combined with the photothermal response of the cotton topping,the feasibility of laser topping was verified both theoretically and experimentally.The laser-topping scheme proposed in this study exhibits high efficiency,environmental protection,and safety,as well as good application prospects.
