The rise of urban traffic flow highlights the growing importance of traffic safety.In order to reduce the occurrence rate of traffic accidents,and improve front vision information of vehicle drivers,the method to impr...The rise of urban traffic flow highlights the growing importance of traffic safety.In order to reduce the occurrence rate of traffic accidents,and improve front vision information of vehicle drivers,the method to improve visual information of the vehicle driver in low visibility conditions is put forward based on infrared and visible image fusion technique.The wavelet image confusion algorithm is adopted to decompose the image into low-frequency approximation components and high-frequency detail components.Low-frequency component contains information representing gray value differences.High-frequency component contains the detail information of the image,which is frequently represented by gray standard deviation to assess image quality.To extract feature information of low-frequency component and high-frequency component with different emphases,different fusion operators are used separately by low-frequency and high-frequency components.In the processing of low-frequency component,the fusion rule of weighted regional energy proportion is adopted to improve the brightness of the image,and the fusion rule of weighted regional proportion of standard deviation is used in all the three high-frequency components to enhance the image contrast.The experiments on image fusion of infrared and visible light demonstrate that this image fusion method can effectively improve the image brightness and contrast,and it is suitable for vision enhancement of the low-visibility images.展开更多
In this paper, a novel fusion framework is proposed for night-vision applications such as pedestrian recognition,vehicle navigation and surveillance. The underlying concept is to combine low-light visible and infrared...In this paper, a novel fusion framework is proposed for night-vision applications such as pedestrian recognition,vehicle navigation and surveillance. The underlying concept is to combine low-light visible and infrared imagery into a single output to enhance visual perception. The proposed framework is computationally simple since it is only realized in the spatial domain. The core idea is to obtain an initial fused image by averaging all the source images. The initial fused image is then enhanced by selecting the most salient features guided from the root mean square error(RMSE) and fractal dimension of the visual and infrared images to obtain the final fused image.Extensive experiments on different scene imaginary demonstrate that it is consistently superior to the conventional image fusion methods in terms of visual and quantitative evaluations.展开更多
目前大多数视觉与激光融合的同步定位与建图(Simultaneous Localization and Mapping,SLAM)算法在进行位姿状态估计时,通常假设其运行环境为静态。然而,当机器人在包含动态对象的环境中移动时,这些对象的出现会改变图像信息,导致数据关...目前大多数视觉与激光融合的同步定位与建图(Simultaneous Localization and Mapping,SLAM)算法在进行位姿状态估计时,通常假设其运行环境为静态。然而,当机器人在包含动态对象的环境中移动时,这些对象的出现会改变图像信息,导致数据关联错误,从而引发SLAM定位误差。针对此问题,提出一种改进的视觉与激光融合SLAM算法,以提高室内动态环境下的三维建图精度和鲁棒性。该方法结合对极几何约束和欧式聚类,实现对动态对象的精准检测和剔除,减少动态对象对SLAM系统定位和建图的干扰。实验结果表明,相较于传统的融合SLAM方法,改进后的算法在动态环境下的定位精度、地图质量以及鲁棒性方面均有提升,为智能机器人、自动驾驶等领域的动态环境建图提供新的解决方案。展开更多
基金the Science and Technology Development Program of Beijing Municipal Commission of Education (No.KM201010011002)the National College Students'Scientific Research and Entrepreneurial Action Plan(SJ201401011)
文摘The rise of urban traffic flow highlights the growing importance of traffic safety.In order to reduce the occurrence rate of traffic accidents,and improve front vision information of vehicle drivers,the method to improve visual information of the vehicle driver in low visibility conditions is put forward based on infrared and visible image fusion technique.The wavelet image confusion algorithm is adopted to decompose the image into low-frequency approximation components and high-frequency detail components.Low-frequency component contains information representing gray value differences.High-frequency component contains the detail information of the image,which is frequently represented by gray standard deviation to assess image quality.To extract feature information of low-frequency component and high-frequency component with different emphases,different fusion operators are used separately by low-frequency and high-frequency components.In the processing of low-frequency component,the fusion rule of weighted regional energy proportion is adopted to improve the brightness of the image,and the fusion rule of weighted regional proportion of standard deviation is used in all the three high-frequency components to enhance the image contrast.The experiments on image fusion of infrared and visible light demonstrate that this image fusion method can effectively improve the image brightness and contrast,and it is suitable for vision enhancement of the low-visibility images.
基金supported in part by the National Natural Science Foundation of China (61533017,U1501251)
文摘In this paper, a novel fusion framework is proposed for night-vision applications such as pedestrian recognition,vehicle navigation and surveillance. The underlying concept is to combine low-light visible and infrared imagery into a single output to enhance visual perception. The proposed framework is computationally simple since it is only realized in the spatial domain. The core idea is to obtain an initial fused image by averaging all the source images. The initial fused image is then enhanced by selecting the most salient features guided from the root mean square error(RMSE) and fractal dimension of the visual and infrared images to obtain the final fused image.Extensive experiments on different scene imaginary demonstrate that it is consistently superior to the conventional image fusion methods in terms of visual and quantitative evaluations.
文摘目前大多数视觉与激光融合的同步定位与建图(Simultaneous Localization and Mapping,SLAM)算法在进行位姿状态估计时,通常假设其运行环境为静态。然而,当机器人在包含动态对象的环境中移动时,这些对象的出现会改变图像信息,导致数据关联错误,从而引发SLAM定位误差。针对此问题,提出一种改进的视觉与激光融合SLAM算法,以提高室内动态环境下的三维建图精度和鲁棒性。该方法结合对极几何约束和欧式聚类,实现对动态对象的精准检测和剔除,减少动态对象对SLAM系统定位和建图的干扰。实验结果表明,相较于传统的融合SLAM方法,改进后的算法在动态环境下的定位精度、地图质量以及鲁棒性方面均有提升,为智能机器人、自动驾驶等领域的动态环境建图提供新的解决方案。
