Object recognition with co-occurrence histograms and false alarm probability analysis for choosing optimal object recognition process parameters

A system for determining optimal search parameters for use in an object recognition procedure designed to find an object in a search image, comprising:

• a general purpose computing device;
• a computer program comprising program modules executable by the computing device, wherein the computing device is directed by the program modules of the computer program to,
• capture model images of the object from a plurality of viewpoints around the object;
• compute a co-occurrence histogram (CH) for each model image, wherein a model image CH is computed by generating counts of every pair of pixels whose pixels exhibit a prescribed pixel characteristic that fall within the same combination of a series of pixel characteristic ranges and which are separated by a distance falling within the same one of a series of distance ranges;
• generate a plurality of search windows each comprising a portion of the search image;
• compute a CH for each search window, wherein a search window CH is computed by generating counts of every pair of pixels whose pixels exhibit a prescribed pixel characteristic that fall within the same combination of said series of pixel characteristic ranges and which are separated by a distance falling within the same one of said series of distance ranges;
• assess a degree of similarity between each model image CH and each of the search window CHs;
• and
• designate a search window associated with a search window CH having a degree of similarity to one of the model image CHs which exceeds a prescribed search threshold as potentially containing the object being sought;
• and wherein,
• said series of pixel characteristic ranges, said series of distance ranges, and the size of the search windows constitute search parameters which are optimized by ensuring these parameters are large enough to minimize the processing required to compute arid assess the similarity of the model image and search window CHs, while at the same time producing an acceptable risk of a false designation that a search window potentially contains the object being sought.

A computer-readable memory for determining optimal search parameters for use in an object recognition procedure designed to find an object in a search image, comprising:

• a general purpose computing device;
• a computer program comprising program modules executable by the computing device, wherein the computing device is directed by the program modules of the computer program to,
• capture model images of the object from a plurality of viewpoints around the object;
• compute a co-occurrence histogram (CH) for each model image, wherein a model image CH is computed by generating counts of every pair of pixels whose pixels exhibit a prescribed pixel characteristic that fall within the same combination of a series of pixel characteristic ranges and which are separated by a distance falling within the same one of a series of distance ranges;
• generate a plurality of search windows each comprising a portion of the search image;
• compute a CH for each search window, wherein a search window CH is computed by generating counts of every pair of pixels whose pixels exhibit a prescribed pixel characteristic that fall within the same combination of said series of pixel characteristic ranges and which are separated by a distance falling within the same one of said series of distance ranges;
• assess a degree of similarity between each model image CH and each of the search window CHs;
• and
• designate a search window associated with a search window CH having a degree of similarity to one of the model image CHs which exceeds a prescribed search threshold as potentially containing the object being sought;
• and wherein,
• said series of pixel characteristic ranges, said series of distance ranges, and the size of the search windows constitute search parameters which are optimized by ensuring these parameters are large enough to minimize the processing required to compute and assess the similarity of the model image and search window CHs, while at the same time producing an acceptable risk of a false designation that a search window potentially contains the object being sought.