Integrating bilingual searches for junk image filtering

To filter out the junk images more effectively, bilingual image search results from two image search engines are integrated to identify the clusters for the junk images and the clusters for the relevant images. Experiments are performed on two image search engines (Google Images in English and Baidu Images in Chinese) by using a large number of bilingual keyword-based queries (5000 bilingual query terms), and our experimental results have shown that integrating bilingual image search results can filter out the junk images effectively.