Modified local tetra pattern based CBIR system

Content based image retrieval (CBIR) is a method of retrieving images from large image database, which has been found according to user’s interest. This paper represents high order local pattern descriptors. Image features are extracted using these descriptors. Image retrieval and indexing algorithm uses these local patterns for content-based image retrieval (CBIR). The local binary pattern (LBP) and local ternary pattern (LTP) encode the relationship between the referenced pixel and its neighbors by calculating graylevel difference. It can extract more detailed information than first order LBP. The local tetra pattern (LTrP) encodes the relationship between the referenced pixel and its surrounding neighbors, based on the directions that are calculated using the first-order derivatives in horizontal and vertical directions. To get more effective result we proposed new method that is modified local tetra pattern (MLTrP). It encodes relationship between pixel based on the directions which are calculated using first order derivatives in horizontal, vertical and diagonal direction. This will improve the performance of the system in terms of average precision and average recall.