Analyses of multiple evidence combination
Proceedings of the 20th annual international ACM SIGIR conference on Research and development in information retrieval
Efficient crawling through URL ordering
WWW7 Proceedings of the seventh international conference on World Wide Web 7
Combining Textual and Visual Cues for Content-Based Image Retrieval on the World Wide Web
CBAIVL '98 Proceedings of the IEEE Workshop on Content - Based Access of Image and Video Libraries
Hi-index | 0.00 |
This paper describes the design, implementation, and evaluation of a novel Web image search engine software. The hardware resources required by our engine are only one commodity personal computer (PC for short) and only one Internet connection with a few Mbps. For a given query and a given time interval, our engine retrieves Web pages relevant to the query during the time interval. Then, our engine analyzes the collected Web pages and finally enumerates images in the Web pages in the descending order of score for the query. We evaluated our engine on 2.4GHz Intel Pentium 4 PC with 512MB RAM and 4.28Mbps effective bandwidth to the Internet. Our experiments show that due to our sophisticated Web crawling algorithm, image discard algorithm, and image scoring algorithm, our engine can typically select 349 images of 2205 images in 192 Web pages in 33.38 seconds.