System and method for classifying a skin infection

A computer-implemented method for classifying skin disorders using an image of a skin portion, via program instructions stored in a memory and executed by a processor, the computer-implemented method comprising:

• receiving the image of the skin portion from an electronic communication device;
• processing the received image of the skin portion, wherein the processing comprises at least one of:
• resizing, removing artifacts and noise, applying new color space, sharpening the image and correcting background;
• segmenting the processed image into two or more regions within the processed image;
• scoring each of the two or more regions based on a correlation coefficient and an edge strength parameter,
• wherein the correlation coefficient is determined for at least one pair of the segmented two or more regions based on intensity of at least one pixel in the pair of the segmented two or more regions;
• and
• further wherein the edge strength parameter is determined for each of the two or more regions based on a sum of gradient magnitude pixel values which lie at a boundary of the segmented region and a total number of gradient magnitude pixels of the segmented region;
• identify a region of interest from the two or more regions based on the score of the two or more regions;
• extracting one or more features from the identified region of interest;
• classifying the skin disorder by mapping the one or more extracted features with pre-stored images, wherein the pre-stored images that map with the one or more extracted features are associated with a particular category of skin disorder;
• and
• rendering the classification of the skin disorder on the electronic communication device.

A computer program product for classifying skin disorders using an image of a skin portion, the computer program product comprising:

• a non-transitory computer-readable medium having computer-readable program code stored thereon, the computer-readable program code comprising instructions that when executed by a processor, cause the processor to:
• receive the image of the skin portion from an electronic communication device;
• process the received image of the skin portion, wherein the processing comprises at least one of:
• resizing, removing artifacts and noise, applying new color space, sharpening the image and correcting background;
• segment the processed image into two or more regions within the processed image;
• scoring each of the two or more regions based on a correlation coefficient and an edge strength parameter,
• wherein the correlation coefficient is determined for at least one pair of the segmented two or more regions based on intensity of at least one pixel in the pair of the segmented two or more regions;
• and
• further wherein the edge strength parameter is determined for each of the two or more regions based on a sum of gradient magnitude pixel values which lie at a boundary of the segmented region and a total number of gradient magnitude pixels of the segmented region;
• identify a region of interest from the two or more regions based on the score of the two or more regions;
• extract one or more features from the identified region of interest;
• classify the skin disorder by mapping the one or more extracted features with pre-stored images, wherein the pre-stored images that map with the one or more extracted features are associated with a particular category of skin disorder;
• and
• render the classification of the skin disorder on the electronic communication device.