Abstract: Integrating multiple features content-based image retrieval can overcome the problems of single feature，but how to organize these features and feature representation methods is difficult in image retrieval.A feature weight assignment approach based on relevance feedback of utilizing genetic algorithm is proposed.By utilizing genetic algorithm and feature weight assignment，the semantic gap between the visual feature and semantic feature is reduced，and the best retrieval results are got in short time.The experimental results show that the recall and precision of this proposed approach are better.This approach can get the best retrieval results for different integrating multiple features.

Key words: multiple features, image retrieval, genetic algorithm, feature weight assignment

摘要： 综合多种特征的图像检索可以克服使用单一特征带来的问题，如何组织这些特征使其发挥最佳作用是基于内容图像检索中的一个难点。提出了一种基于遗传算法相关反馈的图像特征权重调整方法，利用遗传算法和特征权重的调整，缩小低层特征和高层语义之间的语义鸿沟，使检索效果在很短的时间内达到最好。实验表明，该方法查全率和查准率较高，对于各种特征的组合，都能达到最佳的检索效果。

关键词: 多特征, 图像检索, 遗传算法, 特征权重调整