乳腺图像的感兴趣区域(region of interest,ROI)检测是计算机辅助诊断乳腺疾病的第一步,检测效果的提升对减小误诊率有重要的作用.传统方法往往提取单独的视觉特征来描述乳腺图像,通过分类的方法找出包含肿块的区域.然而由于乳腺图像内...乳腺图像的感兴趣区域(region of interest,ROI)检测是计算机辅助诊断乳腺疾病的第一步,检测效果的提升对减小误诊率有重要的作用.传统方法往往提取单独的视觉特征来描述乳腺图像,通过分类的方法找出包含肿块的区域.然而由于乳腺图像内容丰富结构复杂,使用单一的底层视觉容易忽视特征间的相互联系.提出基于相关性特征融合的乳腺图像ROI检测框架(multi-cue integration detection,MCID),通过引入乳腺图像的相关性特征,并与乳腺图像局部视觉特征相融合,辅助乳腺图像ROI的检测,以提高检测准确性.乳腺图像ROI检测实验表明,MCID可提高肿块检测的准确性.展开更多
针对车载智能监控的要求,设计一种嵌入式车体移动报警安全监控系统。系统以TMS320DM642芯片为核心,采用简化的Itti模型来提取监控画面的颜色和亮度特征,并检测和分割出感兴趣区域(Regions of Interest)。根据ROI质心相似度的变化来判断...针对车载智能监控的要求,设计一种嵌入式车体移动报警安全监控系统。系统以TMS320DM642芯片为核心,采用简化的Itti模型来提取监控画面的颜色和亮度特征,并检测和分割出感兴趣区域(Regions of Interest)。根据ROI质心相似度的变化来判断车辆状态。实验结果表明:该系统较为准确地检测和分割感兴趣区域,并能对其进行实时地监控;系统对车辆状态的判断和报警可靠性较高。展开更多
A method is proposed to avoid complex computation in finding the region of interest （ROI） in a mammogram. In the method, the true negative region （TNR） definitely containing no microcalcification clusters （MCCs）...A method is proposed to avoid complex computation in finding the region of interest （ROI） in a mammogram. In the method, the true negative region （TNR） definitely containing no microcalcification clusters （MCCs） is screened out, thus obtaining ROIs, The strategy consists of three steps： （1） the mammogram is partitioned into a set of non-overlapping blocks with an equal size, and for each block, five statistical features are computed, （2） negative blocks are screened out by the threshold method through rough analyses, （3） the more accurate analysis is done by the cost-sensitive support vector machine to eliminate the block definitely containing no MCCs, Experimental results on real mammograms show that 81.71% of TNRs can be screened out by the proposed method.展开更多
This paper proposes a method which uses the extended edge analysis to supplement the inaccurate edge information for better vehicle detection during vehicle detection. The extended edge analysis method detects two ver...This paper proposes a method which uses the extended edge analysis to supplement the inaccurate edge information for better vehicle detection during vehicle detection. The extended edge analysis method detects two vertical edge items, which are the borderlines of both sides of the vehicle, by extending the horizontal edges inaccurately due to the illumination or noise existing on the image. The proposed method extracts the horizontal edges with the method of merging edges by using the horizontal edge information inside the Region of Interest （ROI）, which is set up on the pre-processing step. The bottona line is determined by detecting the shadow regions of the vehicle from the extracted hoodzontal edge one. The general width of the vehicle detecting and the extended edge analyzing methods are carried out side by side on the bottom line of the vehicle to determine width of the vehicle. Finally, the finmal vehicle is detected through the verification step. On the road image with conaplicate background, the vehicle detecting method based on the extended edge analysis is more efficient than the existing vehicle detecting method which uses the edge information. The excellence of the proposed vehicle detecting method is confirmed by carrying out the vehicle detecting experiment on the complicate road image.展开更多
文摘乳腺图像的感兴趣区域(region of interest,ROI)检测是计算机辅助诊断乳腺疾病的第一步,检测效果的提升对减小误诊率有重要的作用.传统方法往往提取单独的视觉特征来描述乳腺图像,通过分类的方法找出包含肿块的区域.然而由于乳腺图像内容丰富结构复杂,使用单一的底层视觉容易忽视特征间的相互联系.提出基于相关性特征融合的乳腺图像ROI检测框架(multi-cue integration detection,MCID),通过引入乳腺图像的相关性特征,并与乳腺图像局部视觉特征相融合,辅助乳腺图像ROI的检测,以提高检测准确性.乳腺图像ROI检测实验表明,MCID可提高肿块检测的准确性.
文摘针对车载智能监控的要求,设计一种嵌入式车体移动报警安全监控系统。系统以TMS320DM642芯片为核心,采用简化的Itti模型来提取监控画面的颜色和亮度特征,并检测和分割出感兴趣区域(Regions of Interest)。根据ROI质心相似度的变化来判断车辆状态。实验结果表明:该系统较为准确地检测和分割感兴趣区域,并能对其进行实时地监控;系统对车辆状态的判断和报警可靠性较高。
文摘A method is proposed to avoid complex computation in finding the region of interest （ROI） in a mammogram. In the method, the true negative region （TNR） definitely containing no microcalcification clusters （MCCs） is screened out, thus obtaining ROIs, The strategy consists of three steps： （1） the mammogram is partitioned into a set of non-overlapping blocks with an equal size, and for each block, five statistical features are computed, （2） negative blocks are screened out by the threshold method through rough analyses, （3） the more accurate analysis is done by the cost-sensitive support vector machine to eliminate the block definitely containing no MCCs, Experimental results on real mammograms show that 81.71% of TNRs can be screened out by the proposed method.
基金supported bythe MKE(The Ministry of Knowledge Economy,Korea),the ITRC(Information Technology ResearchCenter)support program(NIPA-2010-(C1090-1021-0010)),the Brain Korea 21 Project in 2010
文摘This paper proposes a method which uses the extended edge analysis to supplement the inaccurate edge information for better vehicle detection during vehicle detection. The extended edge analysis method detects two vertical edge items, which are the borderlines of both sides of the vehicle, by extending the horizontal edges inaccurately due to the illumination or noise existing on the image. The proposed method extracts the horizontal edges with the method of merging edges by using the horizontal edge information inside the Region of Interest （ROI）, which is set up on the pre-processing step. The bottona line is determined by detecting the shadow regions of the vehicle from the extracted hoodzontal edge one. The general width of the vehicle detecting and the extended edge analyzing methods are carried out side by side on the bottom line of the vehicle to determine width of the vehicle. Finally, the finmal vehicle is detected through the verification step. On the road image with conaplicate background, the vehicle detecting method based on the extended edge analysis is more efficient than the existing vehicle detecting method which uses the edge information. The excellence of the proposed vehicle detecting method is confirmed by carrying out the vehicle detecting experiment on the complicate road image.
