Patent awarded for statistical program

STATE COLLEGE, Pa., Oct. 8 (UPI) -- Two Penn State scientists have received a U.S. patent for a statistical program that might someday make it easier to search the Internet for photographs.

Associate Professors Jia Li and James Wang, said their program -- Automatic Linguistic Indexing of Pictures, or ALIPR -- teaches computers to recognize the contents of photographs, rather than by searching for keywords in the surrounding text, which is how most current image-retrieval systems work.

"Our basic approach is to take a large number of photos -- we started with 60,000 photos -- and to manually tag them with a variety of keywords that describe their contents," said Li. She said a statistical model is then built to teach the computer to recognize patterns in color and texture and to assign keywords to new photos that seem to contain similar scenes.

"Eventually, we hope to reverse the process so that a person can use the keywords to search the Web for relevant images," she added.

Wang and Li said the public can participate in improving ALIPR's accuracy by visiting the Web site -- http://www.alipr.com -- and uploading photographs, then evaluating whether the keywords ALIPR uses to describe the photographs are appropriate.