We present a robust connected-component (CC) based method for automatic detection and segmentation of text in real-scene images. This technique can be applied in robot vision, sign recognition, meeting processing and ...We present a robust connected-component (CC) based method for automatic detection and segmentation of text in real-scene images. This technique can be applied in robot vision, sign recognition, meeting processing and video indexing. First, a Non-Linear Niblack method (NLNiblack) is proposed to decompose the image into candidate CCs. Then, all these CCs are fed into a cascade of classifiers trained by Adaboost algorithm. Each classifier in the cascade responds to one feature of the CC. Proposed here are 12 novel features which are insensitive to noise, scale, text orientation and text language. The classifier cascade allows non-text CCs of the image to be rapidly discarded while more computation is spent on promising text-like CCs. The CCs passing through the cascade are considered as text components and are used to form the segmentation result. A prototype system was built, with experimental results proving the effectiveness and efficiency of the proposed method.展开更多
We address the problem of 3D human pose estimation in a single real scene image. Normally, 3D pose estimation from real image needs background subtraction to extract the appropriate features. We do not make such assum...We address the problem of 3D human pose estimation in a single real scene image. Normally, 3D pose estimation from real image needs background subtraction to extract the appropriate features. We do not make such assumption, In this paper, a two-step approach is proposed, first, instead of applying background subtraction to get the segmentation of human, we combine the segmentation with human detection using an ISM-based detector. Then, silhouette feature can be extracted and 3D pose estimation is solved as a regression problem. RVMs and ridge regression method are applied to solve this problem. The results show the robustness and accuracy of our method.展开更多
Vehicle navigation systems are one of the essential tools for automotive intelligence development,playing a crucial role in the process.This study discusses the components,operation principles,classification,and lates...Vehicle navigation systems are one of the essential tools for automotive intelligence development,playing a crucial role in the process.This study discusses the components,operation principles,classification,and latest technological advances of Vehicle navigation systems,aiming to reveal the current state of the latest technological applications of the system in the automotive industry.The study indicates that the core value of vehicle navigation systems lies in precise positioning,enhanced driving safety,intelligent route planning,and other aspects.At present,the market of vehicle navigation systems is witnessing steady growth and faces intense competition from mobile phone navigation.To hold the upper hand in the competition,the industry should utilize policy support from the government,facing up to challenges and seeking solutions to current problems.In the future,the vehicle navigation system should deeply integrate with artificial intelligence(AD),providing diverse,tailored navigation services for customers.These services should cover driving skills,driving habits,etc.Meanwhile,through constant technological innovation,user experience optimization,and the application of deep leaming,the vehicle navigation system is expected to achieve more efficient human-machine interaction and enhanced driving safety and comfortability,thereby improving its competitiveness in the market and tuning it into an indispensable intelligent companion for drivers.展开更多
文摘We present a robust connected-component (CC) based method for automatic detection and segmentation of text in real-scene images. This technique can be applied in robot vision, sign recognition, meeting processing and video indexing. First, a Non-Linear Niblack method (NLNiblack) is proposed to decompose the image into candidate CCs. Then, all these CCs are fed into a cascade of classifiers trained by Adaboost algorithm. Each classifier in the cascade responds to one feature of the CC. Proposed here are 12 novel features which are insensitive to noise, scale, text orientation and text language. The classifier cascade allows non-text CCs of the image to be rapidly discarded while more computation is spent on promising text-like CCs. The CCs passing through the cascade are considered as text components and are used to form the segmentation result. A prototype system was built, with experimental results proving the effectiveness and efficiency of the proposed method.
基金Supported by the National Basic Research Program of China (Grant No.2006CB303103)Key Program of the National Natural Science Foundation of China (Grant No.60833009)
文摘We address the problem of 3D human pose estimation in a single real scene image. Normally, 3D pose estimation from real image needs background subtraction to extract the appropriate features. We do not make such assumption, In this paper, a two-step approach is proposed, first, instead of applying background subtraction to get the segmentation of human, we combine the segmentation with human detection using an ISM-based detector. Then, silhouette feature can be extracted and 3D pose estimation is solved as a regression problem. RVMs and ridge regression method are applied to solve this problem. The results show the robustness and accuracy of our method.
文摘Vehicle navigation systems are one of the essential tools for automotive intelligence development,playing a crucial role in the process.This study discusses the components,operation principles,classification,and latest technological advances of Vehicle navigation systems,aiming to reveal the current state of the latest technological applications of the system in the automotive industry.The study indicates that the core value of vehicle navigation systems lies in precise positioning,enhanced driving safety,intelligent route planning,and other aspects.At present,the market of vehicle navigation systems is witnessing steady growth and faces intense competition from mobile phone navigation.To hold the upper hand in the competition,the industry should utilize policy support from the government,facing up to challenges and seeking solutions to current problems.In the future,the vehicle navigation system should deeply integrate with artificial intelligence(AD),providing diverse,tailored navigation services for customers.These services should cover driving skills,driving habits,etc.Meanwhile,through constant technological innovation,user experience optimization,and the application of deep leaming,the vehicle navigation system is expected to achieve more efficient human-machine interaction and enhanced driving safety and comfortability,thereby improving its competitiveness in the market and tuning it into an indispensable intelligent companion for drivers.
