Decision trees for computer-aided detection and classification (CAD/CAC) of mines in sidescan sonar imagery
Decision trees for computer-aided detection and classification (CAD/CAC) of mines in sidescan sonar imagery
复制标题
用于侧扫声纳图像中地雷计算机辅助检测和分类 (CAD/CAC) 的决策树
DOI：
--
复制
发表时间：
2002
期刊：
影响因子：
--
通讯作者：
V. Myers
中科院分区：
文献类型：
--
作者：
V. Myers
直观的训练过程和易于解释的决策树的一个有吸引力的解决方案，计算机辅助检测和分类（CAD/CAC）的侧扫声纳图像中的水雷。使用真实的声纳数据，使用不同的分割方法，即分离图像的高光，阴影和背景区域的决策树归纳过程的性能进行评估。还研究了拖鱼俯仰修正的影响。有人认为，所选择的分割算法对分类精度的影响是最小的目标/杂波歧视的情况下，但进行目标分类时，即选择之间的圆柱体，曼塔和球体类变得更加重要。归纳过程，然后扩展到占改变先验类概率和误分类成本。各种变化的后果范围从高灵敏度相对不敏感的成本，并提出了一个案例，使用ROC（接收机操作特性）分析时，比较CAD/CAC方法。最后，检查从检测到分类阶段的进展。
The intuitive training process and easy interpretability make decision trees an attractive solution for computer-aided detection and classification (CAD/CAC) of underwater mines in sidescan sonar imagery. Using real sonar data, performance of the decision tree induction process is evaluated using different segmentation methods, that is separating the image into highlight, shadow and background regions. The effect of towfish pitch correction is also investigated. It is argued that the impact of the chosen segmentation algorithm on classification accuracy is minimal in the case of target/clutter discrimination but becomes more important when carrying out target classification, i.e. choosing between cylinder, manta and sphere classes. The induction process is then extended to account for changing prior class probabilities and misclassification costs. Various changes are made with consequences ranging from high sensitivity to relative insensitivity to costs, and a case is made for the use of ROC (Receiver Operating Characteristic) analysis when comparing CAD/CAC methods. Finally, the progression from detection to classification stages is examined.