Internet search engine major Yahoo has secured an Indian patent for an invention regarding a method for processing image data for the purpose of classifying the images into either graph or photograph.
The patent application for the invention, titled ‘Method of processing image data for purposes of classifying an image represented by the image data’, was filed by Yahoo in 2005 at the Kolkata patent office.
Nirmalya Sinha, deputy controller of patents & designs, Kolkata, granted the patent after being satisfied with the amended claims which met the requirements of the objections raised in the first examination report (FER).
According to a patent document filed by the company, the invention is directed towards a method of processing image data by first processing the colour image data to generate a gray-scale image data, determining measures based on the frequency of occurrence of each gray-scale value in the gray-scale image data, and classifying the colour image data as a type of image data based on the measures. Yahoo submitted that the invention consists of a method and apparatus for processing image data. The colour image data is processed to generate gray scale image data, which may comprise a set of data values. Each data value, of the set of data values, may identify a gray-scale value.
Further, a frequency distribution that indicates a frequency of occurrence of each gray-scale value, is determined, following which the uniformity of the frequency distribution is also detected. Yahoo, one of the major players in the domains of maps and internet search activities, submitted that different types of digital images may satisfy a keyword search. For example, a map of England and photograph of Parliament may both satisfy the keyword search if both were associated with the term England.
Since a map is a graphic, not a photograph, if all the photographs in the set of digital images being searched were eliminated from consideration, then the search for map of England maybe performed more quickly and the result may produce more pertinent results.
Also, numerous situations exist where a system treats a photograph differently from a graphic and thus it would be desirable for the system to know whether a particular digital image is photograph or a graphic. For instance, certain types of digital processing are only applicable only to photographs. Thus, it would be advantageous to determine whether a particular image is a photograph or a graphic because certain types of digital processing may not need to be performed if the digital image is a graphic, Yahoo submitted.