arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1101.0384cs.CV
收藏

基于组合神经网络的皮肤检测

Combining Neural Networks for Skin Detection

School of Engineering and Information Technology, Universiti Malaysia Sabah(马来西亚沙巴大学工程与信息技术学院);Computer and Systems Sciences, Graduate School of Engineering, Osaka Prefecture University(大阪府立大学工程研究生院计算机与系统科学系)

Chelsia Amy Doukim, Jamal Ahmad Dargham, Ali Chekima, Sigeru Omatu

摘要:

评估了两种类型的组合策略,即组合皮肤特征和组合皮肤分类器。应用了几种组合规则,其中使用诸如AND和OR运算符、“投票”、“权重和”的二元运算符和新的神经网络来组合皮肤分类器的输出。从YCbCr颜色空间中选择在其单个特征MLP上给出最高正确检测的三个色度分量作为组合参数。设计MLP神经网络的一个主要问题是在给定一组训练模式的情况下确定最佳的隐层单元数。因此,提出了一种“由粗到细搜索”的方法来寻找隐藏层中的神经元数目。Cb/Cr和Cr特征相结合的策略比Cb-Cr给出的最佳单特征MLP提高了3.01%的正确检测率。与最佳单一特征MLP相比,使用“权重和”规则组合三个皮肤分类器的输出的策略进一步将正确检测提高了4.38%。

英文摘要:

Two types of combining strategies were evaluated namely combining skin features and combining skin classifiers. Several combining rules were applied where the outputs of the skin classifiers are combined using binary operators such as the AND and the OR operators, "Voting", "Sum of Weights" and a new neural network. Three chrominance components from the YCbCr colour space that gave the highest correct detection on their single feature MLP were selected as the combining parameters. A major issue in designing a MLP neural network is to determine the optimal number of hidden units given a set of training patterns. Therefore, a "coarse to fine search" method to find the number of neurons in the hidden layer is proposed. The strategy of combining Cb/Cr and Cr features improved the correct detection by 3.01% compared to the best single feature MLP given by Cb-Cr. The strategy of combining the outputs of three skin classifiers using the "Sum of Weights" rule further improved the correct detection by 4.38% compared to the best single feature MLP.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑