Research On Image Retrieval Algorithm Based On Visual Feature Extraction

The rapid and massive growth of digital images requires effective retrieval methods,which motivates people to research and develop effective image storage,indexing and retrieval technologies.Image retrieval and indexing have been applied in many fields,such as the internet,media,advertising,art,architecture,education,medical,biological and other industries.The text-based image retrieval process first manually labels the image with text,and then uses keywords to retrieve the image.This method of retrieving an image based on the degree of character matching in the image description is time-consuming and subjective.The content-based image retrieval method overcomes the shortcomings of the text-based method,starting from the visual characteristics of the image(color,texture,shape,etc.),and finding similar images in the image library(search range).According to the working principle of general image retrieval,there are three keys to content-based image retrieval:selecting appropriate image features,adopting effective feature extraction methods and accurate feature matching strategies.Based on the research on the key technology of visual feature extraction for image retrieval,the paper proposes four algorithms for image retrieval,Local Color Energy Oriented Pattern and Color Histogram,Stable Interest Point Region,Combination of Dominant Color Descriptor and Hu Moments in Consistent Zone,Combination of region and orientation correlation descriptors.The main research work and contributions of the dissertation are listed as following:1.Local Color Energy Oriented Pattern and Color Histogram is proposed in this paper.Local binary pattern(LBP)is a well-known descriptor that characterizes the image texture.LBP and many LBP-based descriptors perform well on image retrieval.However,they are basically applied to gray-channel of a color image.This paper proposes a multi-channel overall new feature extraction detection algorithm for color images,which is termed as local color energy oriented pattern and color histogram(LCEOPa CH).Oriented by the energy change of the local area of the color image,comparing the intensity change of the reference local pattern with the moved local pattern according to the direction of maximum energy change,LCEOP dynamically represents the local pattern characteristics of the image and can show more information concerning the textural change of the local pattern,comparing with the LBP and LBP-based descriptors that often compare the intensity between the center point in a local neighborhood and the surrounding points.Combining LCEOP features and image color features can better present the image's feature information and achieve superior image retrieval results.The proposed method,LCEOPa CH,has been tested on three data-sets: Corel-1k,Corel-5k,and Corel-10 k,and the experiments demonstrate that it outperforms state-of-the-art algorithms.2.Stable Interest Point Region is proposed in this paper.The retrieval accuracy is reduced by interest points of non-interest region in traditional interest point retrieval methods.Aiming at these deficiencies,a new method called Stable Interest Point Region(SIPR)is proposed for retrieving images from regions based on stable interest points.Firstly,the interest points of the query image and the candidate image are detected.Moreover,the neighborhood gray scale information is extracted and the pseudo Zernike moment is calculated.As a result,the best matching point pair can be found in the process of distances comparing.Secondly,the corresponding convex hull area is calculated by matching point pairs to obtain a stable interest point area.Finally,the color and texture information of the image in the stable interest point region are extracted by the color histogram combined with the Gabor wavelet transform.The experimental results show that the proposed algorithm,SIPR,removes the influence of irrelevant points of interest,as well as overcomes the shortcomings of the traditional interest algorithms which only extract the local features of the edge.Compared with similar algorithms,the accuracy is significantly improved.3.Combination of Dominant Color Descriptor and Hu Moments in Consistent Zone is proposed in this paper.The rapidly increasing number of digital images requires effective retrieval.Meanwhile,the dominant color descriptor has been widely used in image processing.Due to the influence of lighting and other factors,the same color in nature may have some different changes.The human eye is usually more sensitive to zones of consistent color,often identifying objects by zones of consistency.Therefore,A new method called Combination of Dominant Color Descriptor and Hu Moments(CDC-HM)is proposed.It first applies the texton template to detect and extract the consistent zone of an image,and calculates the dominant color descriptor feature on the pixels in this consistent zone.Besides,the translation and rotation invariance of the Hu moments feature is applied to extract the shape information in the same consistent zone of the image.Finally,the combination of the dominant dolor descriptor and the Hu moments is used for content-based image retrieval.The proposed algorithm is tested on three data sets:Corel-1k,Corel-5k and Corel-10 k,and the experimental results show that it is superior to the current content-based image retrieval methods.4.Combination of region and orientation correlation descriptors is proposed in this paper.A large number of growing digital images require retrieval effectively,but the trade-off between accuracy and speed is a tricky problem.This paperwork proposes a lightweight and efficient image retrieval approach by combining region and orientation correlation descriptors(CROCD).The region color correlation pattern and orientation color correlation pattern are extracted by the region descriptor and the orientation descriptor,respectively.The feature vector of the image is extracted from the two correlation patterns.The proposed algorithm has the advantages of statistic and texture description methods,and it can represent the spatial correlation of color and texture.The feature vector has only 80 dimensions for full color images specifically.Therefore,it is very efficient in image retrieving.The proposed algorithm is extensively tested on three datasets in terms of precision and recall.The experimental results demonstrate that the proposed algorithm outperforms other state-of-the-art algorithms.