基于色彩和边缘特征的图像检索方法

Image retrieval method based on color and edge features

为了改善基于内容的图像检索的效果和提高其检索效率,提出一种基于色彩和边缘特征的图像检索方法。首先将RGB图像分成几个子图像的形式,然后,对于每一个子图像提取其色彩特征和边缘特征,其中边缘特征的获得采用了瞬时保持(MP)边缘检测技术。将这两种特征结合在一起使用,可以实现准确快捷的图像检索。实验结果表明,该方法在检索精度和检索效率上都高于Cheng的两种方法且所用的时间分别为Cheng的方法的10%和3%,检索精度提高近20%。

In order to improve the effectiveness and efficiency of the Content-Based Image Retrieval （CBIR）, in this paper, a new method was proposed. Both color feature and edge feature were used in this method. At first, RGB image was departed into several sub-images. Then, color feature and edge feature were extracted from every sub-image, where edge feature was acquired by utilizing the Moment Preserving （MP） edge detection. Combining the above two features, an effective and efficient image retrieval method was obtained. The experimental results show that the proposed method has higher precision and faster speed than Cheng＇s two methods. Compared with Cheng＇s two methods, retrieval time of the method proposed in this naner is its 10% and 3%. and retrieval precision has been improved bv 20%.