英语高手近来帮个忙,最好是学图像的!帮忙翻译一下,不要用翻译工具粘啊

来源:百度知道 编辑:UC知道 时间:2024/05/28 12:44:09
本算法是在系统分析研究相关文献基础上,对人脸分割各种算法的融合和改进,适合于复杂背景下背景前景稳定的情况下进行人脸定位分割。在经过大量实验测试后,我们发现在对目标图像进行肤色检测时,如果背景中存在大量与肤色相近的色彩区域时,这些区域也被分割出来作为目标区域,进行人脸区域分割时,存在一定的误检率,步骤2中人脸模型的建立很好的解决了这一问题,使得我们能更精确的勾勒目标人脸。
经过一定的测试,本算法综合适用性强,误检率低,具有很强的可行性。
怎么没人来回答啊?

Analysis of the algorithm is at the basis of relevant literature on research on a variety of Human Face Segmentation Algorithm for the integration and improvement of fit in the context of the complexity of the context of the prospects for the stability of the partition face positioning. After a large number of experimental tests, we found the target in the detection of color images, if there is substantial background color with the color similar to the region, these regions have been split out as a target region, the regional division of the face to people, there must the false rate, step 2 model of the human face of a very good solution to this problem so that we can outline a more precise goal face.
After some testing, General applicability of this algorithm, the low rate of false, has a strong feasibility.