Advanced image pattern searching technic
Abstract : With many potential practical applications, content-based image retrieval (CBIR) has attracted substantial attention during the past few years. Hence the authors implemented an advanced image pattern searching algorithm. In the proposed system, the images are searched based on the image patterns such as pixels, resolution, height and width of the image. To carry out the image searching process semi supervising algorithm is used. In order to carry out the search among multiple images, the parameters such as pixel height and width, resolution of the input image was compared with the images in the database. If there is a match between the two images, the matched image will be extracted and displayed as output. On selecting this extracted image one can add hidden text character, gray scale and color change property.
? In the existing system, images are searched by the text or by the image which is given in the input field.
? Based on the image or text which is given in the input field, images are displayed in the output.
? Searching was carried out based on text search and the images were listed out.
? The approach was very slow, time consuming and less efficient. On considering all these issues, the authors proposed a new method based on pixel matching.
? In order to overcome the issues of the existing method, the proposed method was organized to provide high efficiency.
? But even then there exists various drawbacks in carrying out the process. On considering all these, the authors proposed an approach based on pixel matching, following a semi supervised learning approach.
? Some algorithms have a time complexityproblem. So, there is a need to develop an effective and accurate image forgery detection algorithm.
? They also discussed the main challenges in that field and how other researchers addressed those challenges.
• In order to overcome the issues of the existing method, the proposed method was organized to provide high efficiency.
• Pixel based searching algorithm was implemented that considers the resolution, pixel height and width.
• The approach follows a semi supervised learning algorithm. Once the image was extracted, certain property such as, text hiding, gray scale and color changing property was imposed on the image to differentiate the duplicate property of the image.
• The proposed approach was found to be highly efficient, fast and accurate.
? The main goal of the authors was to analyze the performance of their system by combining both supervised and unsupervised methods.
? Their experimental results proved that their proposed approachachieved better performance in the aspects of both efficiency and accuracy compared with the state-of-the-art approaches.
? They further demonstrated its practical performance with several challenging forgery images created with state-of-the-art tools.
? The robustness of the SIFT features with regards to local transforms renders that this method was able to detect general region duplications with efficient computation.
|