This paper suggests a new algorithm to solve problems of the current retinex algorithm such as distortion of grey out and color noise due to the individual treatment of RGB channel and log function,and halo effect occ...This paper suggests a new algorithm to solve problems of the current retinex algorithm such as distortion of grey out and color noise due to the individual treatment of RGB channel and log function,and halo effect occurred by use of the Gaussian filter.The current retinex algorithm treats each channel in RGB space that brings a phenomenon to change the rate of RGB.To improve this phenomenon,the color information was fixed in the HSV color space,and retinex treatment was conducted against the V value,a luminance feature.Linear treatment was carried out to remove color noise occurred by the use of log function.S value,a saturation value was compensated in proportion to the change of V value in luminance to obtain a clearer image.The proposed algorithm was evaluated against the landscape images that had strong backlit phenomena,and it is proved to have a better performance than the current retinex algorithm,multiscale retinex with cdor restoration(MSRCR).展开更多
嵌入式设备中部署深度学习检测模型往往面临算力不足的问题,而感兴趣区域(ROI)提取可作为一种高效的性能优化手段。文章提出一种基于HSV(Hue,Saturation,Value)色彩空间模型的ROI提取的方法,将检测目标的像素信息转化到HSV色彩空间,在色...嵌入式设备中部署深度学习检测模型往往面临算力不足的问题,而感兴趣区域(ROI)提取可作为一种高效的性能优化手段。文章提出一种基于HSV(Hue,Saturation,Value)色彩空间模型的ROI提取的方法,将检测目标的像素信息转化到HSV色彩空间,在色相-饱和度(H-S)平面引入DBSCAN(Density-Based Spatial Clustering of Applications with Noise)聚类算法,精确定位目标的主色彩像素在H-S平面上的分布位置,同时过滤杂乱色彩,然后通过Quickhull(快壳)凸包算法,从散点数据中拟合出主色彩的精确分布范围。根据获取的主色彩范围对像素进行遍历,可以根据色彩信息有效地提取ROI。实验结果表明,经过该方法优化后的Faster R-CNN(Faster Regions with Convolutional Neural Networks)算法,较原模型减少了57.08%的平均推理耗时,同时精确率提升了0.9百分点。这对于嵌入式设备中进行实时目标检测具有重要的现实意义。展开更多
Machine vision has been recently utilized for quality control of food and agricultural products, which was traditionally done by manual inspection. The present study was an attempt for automatic defect detection and s...Machine vision has been recently utilized for quality control of food and agricultural products, which was traditionally done by manual inspection. The present study was an attempt for automatic defect detection and sorting of some single-color fruits such as banana and plum. Fruit images were captured using a color digital camera with capturing direction of zero degree and under illuminant D65. It was observed that growing decay and time-aging made surface color changes in bruised parts of the object. 3D RGB and HSV color vectors as well as a single channel like H （hue）, S （saturation）, V （value） and grey scale images were applied for color quantization of the object. Results showed that there was a distinct threshold in the histogram of the S channel of images which can be applied to separate the object from its background. Moreover, the color change via the defect and time-aging is correctly distinguishable in the hue channel image. The effect of illumination, gloss and shadow of 3D image processing is less noticeable for hue data in comparison to saturation and value. The value of H channel was quantized to five groups based on the difference between each pixel value and the H value of a healthy object. The percentage of different degree of defects can be computed and used for grading the fruits.展开更多
An adaptive background model based on max-imum statistical probability and a shadow suppression scheme for indoor and outdoor people detection by exploiting hue saturation value(HSV)color information is proposed.To ob...An adaptive background model based on max-imum statistical probability and a shadow suppression scheme for indoor and outdoor people detection by exploiting hue saturation value(HSV)color information is proposed.To obtain the initial background scene,the frequency of R,G,and B component values for each pixel at the same position in the learning sequence are respec-tively calculated;the R,G,and B component values with the biggest ratios are incorporated to model the initial background.The background maintenance,or the so-called background re-initiation,is also proposed to adapt to scene changes such as illumination changes and scene geometry changes.Moving cast shadows generally exhibit a challenge for accurate moving target detection.Based on the observation that a shadow cast on a background region lowers its brightness but does not change its chro-maticity significantly,we address this problem in the ar-ticle by exploiting HSV color information.In addition,quantitative metrics is introduced to evaluate the algo-rithm on a benchmark suite of indoor and outdoor video sequences.The experimental results are given to show the performance of the algorithm.展开更多
针对复杂背景条件下目标难以识别的问题,采用彩色偏振成像技术,提出了一种基于彩色偏振图像的目标增强方法。该方法首先根据分焦平面彩色偏振相机获得的数据得到彩色线偏振度(degree of linear polarization, DoLP)、彩色偏振角(angle o...针对复杂背景条件下目标难以识别的问题,采用彩色偏振成像技术,提出了一种基于彩色偏振图像的目标增强方法。该方法首先根据分焦平面彩色偏振相机获得的数据得到彩色线偏振度(degree of linear polarization, DoLP)、彩色偏振角(angle of polarization, AoP)和彩色强度(S0)图像;然后利用目标和背景的彩色偏振特性差异大的特点提取DoLP、AoP和S0的视觉显著度,使目标得到初步的增强;随后将3种视觉显著度图像转到HSV空间进行融合,最后转到RGB空间显示。使用对比度和矢量角度距离作为客观评价指标开展实验,多个实验场景数据表明,融合图像的对比度和矢量角度距离分别比融合前图像最高提升了3.971倍和1.711倍。展开更多
基金The Brain Korea21Project in 2011 andthe MKE(the Ministry of Knowledge Economy),Korea,under the ITRC(Information Technology Research Center)support programsupervised by the NIPA(National IT Industry Promotion Agency)(NIPA-2011-C1090-1121-0010)
文摘This paper suggests a new algorithm to solve problems of the current retinex algorithm such as distortion of grey out and color noise due to the individual treatment of RGB channel and log function,and halo effect occurred by use of the Gaussian filter.The current retinex algorithm treats each channel in RGB space that brings a phenomenon to change the rate of RGB.To improve this phenomenon,the color information was fixed in the HSV color space,and retinex treatment was conducted against the V value,a luminance feature.Linear treatment was carried out to remove color noise occurred by the use of log function.S value,a saturation value was compensated in proportion to the change of V value in luminance to obtain a clearer image.The proposed algorithm was evaluated against the landscape images that had strong backlit phenomena,and it is proved to have a better performance than the current retinex algorithm,multiscale retinex with cdor restoration(MSRCR).
文摘嵌入式设备中部署深度学习检测模型往往面临算力不足的问题,而感兴趣区域(ROI)提取可作为一种高效的性能优化手段。文章提出一种基于HSV(Hue,Saturation,Value)色彩空间模型的ROI提取的方法,将检测目标的像素信息转化到HSV色彩空间,在色相-饱和度(H-S)平面引入DBSCAN(Density-Based Spatial Clustering of Applications with Noise)聚类算法,精确定位目标的主色彩像素在H-S平面上的分布位置,同时过滤杂乱色彩,然后通过Quickhull(快壳)凸包算法,从散点数据中拟合出主色彩的精确分布范围。根据获取的主色彩范围对像素进行遍历,可以根据色彩信息有效地提取ROI。实验结果表明,经过该方法优化后的Faster R-CNN(Faster Regions with Convolutional Neural Networks)算法,较原模型减少了57.08%的平均推理耗时,同时精确率提升了0.9百分点。这对于嵌入式设备中进行实时目标检测具有重要的现实意义。
文摘Machine vision has been recently utilized for quality control of food and agricultural products, which was traditionally done by manual inspection. The present study was an attempt for automatic defect detection and sorting of some single-color fruits such as banana and plum. Fruit images were captured using a color digital camera with capturing direction of zero degree and under illuminant D65. It was observed that growing decay and time-aging made surface color changes in bruised parts of the object. 3D RGB and HSV color vectors as well as a single channel like H （hue）, S （saturation）, V （value） and grey scale images were applied for color quantization of the object. Results showed that there was a distinct threshold in the histogram of the S channel of images which can be applied to separate the object from its background. Moreover, the color change via the defect and time-aging is correctly distinguishable in the hue channel image. The effect of illumination, gloss and shadow of 3D image processing is less noticeable for hue data in comparison to saturation and value. The value of H channel was quantized to five groups based on the difference between each pixel value and the H value of a healthy object. The percentage of different degree of defects can be computed and used for grading the fruits.
文摘An adaptive background model based on max-imum statistical probability and a shadow suppression scheme for indoor and outdoor people detection by exploiting hue saturation value(HSV)color information is proposed.To obtain the initial background scene,the frequency of R,G,and B component values for each pixel at the same position in the learning sequence are respec-tively calculated;the R,G,and B component values with the biggest ratios are incorporated to model the initial background.The background maintenance,or the so-called background re-initiation,is also proposed to adapt to scene changes such as illumination changes and scene geometry changes.Moving cast shadows generally exhibit a challenge for accurate moving target detection.Based on the observation that a shadow cast on a background region lowers its brightness but does not change its chro-maticity significantly,we address this problem in the ar-ticle by exploiting HSV color information.In addition,quantitative metrics is introduced to evaluate the algo-rithm on a benchmark suite of indoor and outdoor video sequences.The experimental results are given to show the performance of the algorithm.
文摘针对复杂背景条件下目标难以识别的问题,采用彩色偏振成像技术,提出了一种基于彩色偏振图像的目标增强方法。该方法首先根据分焦平面彩色偏振相机获得的数据得到彩色线偏振度(degree of linear polarization, DoLP)、彩色偏振角(angle of polarization, AoP)和彩色强度(S0)图像;然后利用目标和背景的彩色偏振特性差异大的特点提取DoLP、AoP和S0的视觉显著度,使目标得到初步的增强;随后将3种视觉显著度图像转到HSV空间进行融合,最后转到RGB空间显示。使用对比度和矢量角度距离作为客观评价指标开展实验,多个实验场景数据表明,融合图像的对比度和矢量角度距离分别比融合前图像最高提升了3.971倍和1.711倍。
