目标分类是计算机视觉研究中的重要基本问题之一.为提高目标分类的准确率,由对目标进行人工分类的完整过程所得到的启发,提出了一种视觉注意力模型与CNN相结合的目标分类新方法.该方法与传统目标分类方法相比,在分类过程上更接近于人工行为,有明显的生物学优势.首先,建立一个基于分类任务的眼动数据库,研究并记录人在进行目标分类时的视觉行为;然后,利用该数据库训练出一个结合低层特征(如方向、颜色、强度等)及高层特征(如人、脸、汽车等)的视觉注意力模型,以此,预测人工区分不同目标时所感兴趣的区域;最后设计CNN网络,利用视觉注意力模型所得到的感兴趣区域进行目标分类.实验结果表明,所提出的视觉注意力模型可以更准确地预测人在分类时的感兴趣区域,可显著提高目标分类的准确度,并且收敛速度更快.%Objects classification is one of the most significant problems in computer vision. For improving the accu-racy of objects classification,we put forward a new classification method enlightened the whole process that human distinguish different types of objects. Our method mixed visual saliency model and CNN, is more close to human and has apparently biological advantages. Firstly,we built an eye-tracking database to learn people visual behaviors when they classify various objects and recorded the eye-tracking data. Secondly, this database is used to train a learning-based visual attention model,which is based on low-level(e.g., orientation, color, intensity, etc.)and high-level(e.g.,faces,people,cars,etc.)image features to analyze and predict the human's classification RoIs. Finally,we established a CNN framework to classify RoIs. The results of the experiment showed our attention model can determine saliency regions and predict human's classification RoIs more precisely and our classification method improved the efficiency of classification markedly.
展开▼
