BLOCK-BASED IMAGE EXPOSURE ASSESSMENT AND INDOOR/OUTDOOR CLASSIFICATION
С. С. Ефимов, Alexey Nefyodov, M. Rychagov · 2007
Automatic image quality assessment as well as image classification onto indoor/outdoor ones are performed by subdivision of image under analysis onto a number of fragments. Detecting overexposed, underexposed, low contrast or normal images is carried out by analyzing luminance histogram of the image and its constituting segments. At the stage of image classification, support vector machines (SVMs) are used for color features processing for each fragment of the image and wavelet analysis is used for texture description.