用于场景分类的显著建筑物区域检测

为了实现城市区域内室外场景的快速分类，提出了一种基于视觉注意力选择机制的显著建筑物区域检测方法.该方法首先通过Gabor算子对图像进行滤波，得到水平方向与垂直方向的联合增强图像，然后利用基于相位傅里叶变换(Phase Fourier Transform, PFT)的显著性映射算法提取视野中的显著建筑物候选区域，最后通过方向直方图判别算法去除干扰目标，并采用CV (Chan Vese)模型实现显著建筑物区域的分割.该方法在注意力选择机制及建筑物方向特征的基础上，提取具有场景代表性的建筑物区域信息，增强了算法的时效性和场景分析能力，同时实现了区域信息在像素级上的精确检测，并将其应用于美国南加州大学的多类室外场景数据库.实验结果表明，显著建筑物区域检测方法在检测结果、计算速度和鲁棒性等方面均取得了较为满意的效果.

Salient Building Region Detection for Scene Classification

In order to achieve outdoor scene classification in urban areas quickly,an effective method for detecting salient building region based on visual attention selection mechanism was proposed.Firstly,eight-directional Gabor filter is used to obtain horizontal-vertical enhanced image.Secondly,salient building candidate regions are extracted through mapping from salient map which is constructed by phase Fourier tranform to the enhanced image.Finally,an orientation histogram discrimination algorithm is used to el...