Combining convolutional neural networks and slantlet transform for an effective image retrieval scheme

<p>In the latest years there has been a profound evolution in computer science and technology, which incorporated several fields. Under this evolution, Content Base Image Retrieval (CBIR) is among the image processing field. There are several image retrieval methods that can easily extract feature as a result of the image retrieval methods’ progresses. To the researchers, finding resourceful image retrieval devices has therefore become an extensive area of concern. Image retrieval technique refers to a system used to search and retrieve images from digital images’ huge database. In this paper, the author focuses on recommendation of a fresh method for retrieving image. For multi presentation of image in Convolutional Neural Network (CNN), Convolutional Neural Network - Slanlet Transform (CNN-SLT) model uses Slanlet Transform (SLT). The CBIR system was therefore inspected and the outcomes benchmarked. The results clearly illustrate that generally, the recommended technique outdid the rest with accuracy of 89 percent out of the three datasets that were applied in our experiments. This remarkable performance clearly illustrated that the CNN-SLT method worked well for all three datasets, where the previous phase (CNN) and the successive phase (CNN-SLT) harmoniously worked together.</p>

CONTENT-BASED IMAGE RETRIEVAL USING EXPRESSION SENSITIVITY BY FUZZY INFERENCE SYSTEM

Image retrieval can be divided into two types context-based and the content-based. Image retrieval based on the content refers to the image features such as color, texture, shape, semantics or sensations. This paper addresses the content-base image retrieval system based on expression sensitivity. It can be image or text query for input the system. Based on Itten theory, expression sensitivity consist of warm, cold, relax, anxious, and life. The research system uses two fuzzy inference system. Firstly, fuzzy inference system is used to decide image region of color. The image size is 256 x 256 pixel. Output the first fuzzy inference system is input for the second fuzzy inference system. The second fuzzy inference system is used to determined expression sensitivity of image. Degree of accuracy based on respondent from 50 images and 20 respondents is 42% for text query and 55% for image query. The further research, it can be used for other image such as medical image with certain criteria.

Image Retrieval Using Mutual Information

In this paper, we study an information theoretic approach to image similarity measurement for content-base image retrieval. In this novel scheme, similarities are measured by the amount of information the images contained about one another mutual information (MI). The given approach is based on the premise that two similar images should have high mutual information, or equivalently, the querying image should convey high information about those similar to it. The method first generates a set of statistically representative visual patterns and uses the distributions of these patterns as images content descriptors. To measure the similarity of two images, we develop a method to compute the mutual information between their content descriptors. Two images with larger descriptor mutual information are regarded as more similar. We present experimental results, which demonstrate that mutual information is a more effective image similarity measure than those have been used in the literature such as Kullback-Leibler divergence and L2 norms.

Analisis Seleksi Citra Mirip dengan Memanfaatkan Konsep CBIR dan Algoritma Threshold

Content base image retrieval (CBIR) is the concept of image retrieval by comparing the existing image on the sample to that of the database (query by example). CBIR process based on color is carried out using adaptive color histogram concept, while one based on shape is performed using moment concept. Following up the process results, a sorting process is done based on a threshold value of the sample image through the utilization threshold algorithm. The image displayed is be sorted from the one that is nearly similar to the query image (example) to the resemblance of the lowest (aggregation value). The threshold value of the query image used as reference is compared with the aggregation value of the image database. If the comparison in the search for similarities by using the concept of fuzzy logic approaches 1, the comparison between the threshold value and the aggregation value is almost the same. Otherwise, if it reaches 0, the comparison results in a lot of differences.