Performing Content-Based Image Retrieval (CBIR) from Internet databases connected through Peer-to-Peer (P2P) network, abbreviated as P2P-CBIR, helps to effectively explore the large-scale image database distributed over connected peers. Decentralized unstructured P2P framework is adopted in our system to compromise with the structured one while still reserving flexible routing control when peers join/leave or network fails. The P2P-CBIR search engine is designed to provide multi-instance query with multi-feature types to effectively reduce network traffic and maintain high retrieval accuracy. The proposed P2P-CBIR system is also designed to provide scalable retrieval control, which can adaptively control the query scope and progressively refine the accuracy of retrieved results. We also proposed to provide the most updated local database characteristics for the P2P-CBIR users. By reconfiguring system at each regular interval time, it can effectively reduce trivial peer routing and retrieval operations due to imprecise configuration. Experiments demonstrated that the average recall rate of the proposed P2P-CBIR with reconfiguration is higher than the one without about 20%, and the latter outperforms previous methods, i.e., firework query model (FQM) and breadth-first search (BFS) about 20% and 120%, respectively, under the same range of TTL values.

Processing images for specific targets on a large scale has to handle various kinds of contents with regular processing steps. To segment objects in one image, we utilized dual multiScalE Graylevel mOrphological open and close recoNstructions (SEGON) to build a background (BG) gray-level variation mesh, which can help to identify BG and object regions. It was developed from a macroscopic perspective on image BG gray levels and implemented using standard procedures, thus robustly dealing with large-scale database images. The image segmentation capability of existing methods can be exploited by the BG mesh to improve object segmentation accuracy. To evaluate the segmentation accuracy, the probability of coherent segmentation labeling, i.e., the normalized probability random index (PRI), between a computer-segmented image and the hand-labeled one is computed for comparisons. Content-based image retrieval (CBIR) was carried out to evaluate the object segmentation capability in dealing with large-scale database images. Retrieval precision-recall (PR) and rank performances, with and without SEGON, were compared. For multi-instance retrieval with shape feature, AdaBoost was used to select salient common feature elements. For color features, the histogram intersection between two scalable HSV descriptors was calculated, and the mean feature vector was used for multi-instance retrieval. The distance measure for color feature can be adapted when both positive and negative queries are provided. The normalized correlation coefficient of features among query samples was computed to integrate the similarity ranks of different features in order to perform multi-instance with multifeature query. Experiments showed that the proposed object segmentation method outperforms others by 21% in the PRI. Performing SEGON-enabled CBIR on large-scale databases also improves on the PR performance reported elsewhere by up to 42% at a recall rate of 0.5. The proposed object segmentation method can be extended to extract other image features, and new feature types can be incorporated into the algorithm to further improve the image retrieval performance.

Performing Content-Based Image Retrieval (CBIR) on Internet connected databases through Peer-to-Peer (P2P) network (P2P-CBIR) effectively explores the large-scale image database distributed over connected peers. In additional to enlarge the retrieval scale from server-client to P2P networks, the required computation and network traffics for performing CBIR can also be distributed. Decentralized unstructured P2P framework is adopted to perform the P2P-CBIR in considering control flexibility, under which reconfiguration is feasible to improve the retrieval performance. Based on this framework, the proposed P2P-CBIR can progressively refine the accuracy of retrieved images when they are transmitted on the routes toward the query peer. With flexible P2P framework, reconfiguration can be carried out on the P2P-CBIR, which helps to improve retrieval efficiency. Experiments verified that the P2P-CBIR with reconfiguration outperforms the one without, and the latter outperforms server-client approaches in both retrieval accuracy and network traffics.

A simple and regular foreground identification (FGID) method is proposed for image indexing. The gray-level Morphological open/close by reconstruction (MOR/MCR) is operated on one image in a dual and multi-scale approach to construct a background gray-level mesh to distinguish foregrounds (FGs). The highly regular MOR/MCR operations make it feasible to deal with FG segmentation of volume images. The FGID efficiency is verified by the image retrieval performance, i.e., recall, precision and rank.With precisely identified FGs, MPEG-7 shape descriptors, in additional to color ones, can be used to improve the image retrieval performance. For the retrieval unit, a greedy boosting retrieval method is used to perform shape-based multiinstance query in considering the feature element dependency. To perform multi-instance query with multiple features, the retrieval unit integrates the similarity ranks of different feature types according to the feature saliency among query samples to yield the final similarity rank. The normalized correlation coefficient of features among query samples is computed to provide weighting factors for integrating ranks. Experiments show that the FGID unit helps much in improving the retrieval performances, i.e., 7% improvement for the precision-recall (PR) and 20% improvement for the averaged normalized modified retrieval rank (ANMRR), as compared to non-FGID ones.

We proposed to utilize the scalable peer-to-peer network to perform the content-based image retrieval and mining, i.e, P2P-CBIRM. The decentralized unstructured P2P model with certain overheads, i.e., peer clustering and update procedures, is adopted to compromise with the structured one while still reserving flexible routing control when peers join/leave or network fails. The peer CBIRM engine is designed to utilize multi-instance query with multi-feature types to effectively reduce network traffic while maintaining high retrieval accuracy. It helps to enhance the knowledge discovery and image data mining capability. The proposed P2P-CBIRM system provides the scalable retrieval and mining function that the query scope and retrieval accuracy can be adaptively and progressively controlled. To improve the query efficiency (recall-rate/query-scope), it effectively utilizes both: 1) forwarding query message (forward phase) to reduce the query scope and 2) transmitting retrieval results (backward phase) such that activated peers keep filtering high similarity images on the link-path toward the query peer. Experiments show that the query efficiency of the scalable retrieval approach is better than previous methods, i.e., firework query model and breadth-first search. It provides a scalable knowledge discovery platform for efficient image data mining applications. We also proposed to optimally configure the P2P-CBIRM system such that, under a certain number of online users, it would yield the highest recall rate. Simulations demonstrate that, with the optimal configuration, recall rates can be improved to 2.5 to 3 times larger while the network traffic of each peer is reduced to 30% of the original, under the same number of on-line users.