Image Similarity

Abstract: Advances in data storage and image acquisition technologies have enabled the creation of large image datasets. In this scenario, it is necessary to develop appropriate information systems to efficiently manage these collections. The commonest approaches use the so-called Content-Based Image Retrieval (CBIR) systems. Basically, these systems try to retrieve images similar to a user-defined specification or pattern (eg, shape sketch, image example). Their goal is to support image retrieval based on content properties (eg, ...

Diagnostic radiology requires accurate interpretation of complex signals in medical images. Content-based image retrieval (CBIR) techniques could be valuable to radiologists in assessing medical images by identifying similar images in large archives that could assist with decision support. Many advances have occurred in CBIR, and a variety of systems have appeared in nonmedical domains; however, permeation of these methods into radiology has been limited. Our goal in this review is to survey CBIR methods and systems from the perspective of application to radiology and to identify approaches developed in nonmedical applications that could be translated to radiology. Radiology images pose specific challenges compared with images in the consumer domain; they contain varied, rich, and often subtle features that need to be recognized in assessing image similarity. Radiology images also provide rich opportunities for CBIR: rich metadata about image semantics are provided by radiologists, and this information is not yet being used to its fullest advantage in CBIR systems. By integrating pixel-based and metadata-based image feature analysis, substantial advances of CBIR in medicine could ensue, with CBIR systems becoming an important tool in radiology practice.

Assessment of image similarity is fundamentally important to numerous multimedia applications. The goal of similarity assessment is to automatically assess the similarities among images in a perceptually consistent manner. In this paper, we interpret the image similarity assessment problem as an information fidelity problem. More specifically, we propose a feature-based approach to quantify the information that is present in a reference image and how much of this information can be extracted from a test image to assess the similarity between the two images. Here, we extract the feature points and their descriptors from an image, followed by learning the dictionary/basis for the descriptors in order to interpret the information present in this image. Then, we formulate the problem of the image similarity assessment in terms of sparse representation. To evaluate the applicability of the proposed feature-based sparse representation for image similarity assessment (FSRISA) technique, we apply FSRISA to three popular applications, namely, image copy detection, retrieval, and recognition by properly formulating them to sparse representation problems. Promising results have been obtained through simulations conducted on several public datasets, including the Stirmark benchmark, Corel-1000, COIL-20, COIL-100, and Caltech-101 datasets.

A hierarchical algorithm is presented for determining the similarity and equivalence of document images. Features extracted from the CCITT fax compressed representations of two images are compared to determine their visual similarity and whether they are equivalent. Pass codes in the compressed data are used as features. A fixed grid is imposed on the image and a feature vector is derived from the number of pass codes in each grid cell. The feature vectors are compared to locate a group of documents that are visually similar to the input image. The equivalence of two documents is determined by applying the Hausdorff distance to the two dimensional arrangement of pass codes in small patches of each image

This article describes the use of a medical image retrieval system on a database of 16'000 fractures, selected from surgical routine over several years. Image retrieval has been a very active domain of research for several years. It was frequently proposed for the medical domain, but only few running systems were ever tested in clinical routine. For the planning of surgical interventions after fractures, x-ray images play an important role. The fractures are classified according to exact fracture location, plus whether and to which degree the fracture is damaging articulations to see how complicated a reparation will be. Several classification systems for fractures exist and the classification plus the experience of the surgeon lead in the end to the choice of surgical technique (screw, metal plate, ...). This choice is strongly influenced by the experience and knowledge of the surgeons with respect to a certain technique. Goal of this article is to describe a prototype that supplies similar cases to an example to help treatment planning and find the most appropriate technique for a surgical intervention. Our database contains over 16'000 fracture images before and after a surgical intervention. We use an image retrieval system (GNU Image Finding Tool, GIFT) to find cases/images similar to an example case currently under observation. Problems encountered are varying illumination of images as well as strong anatomic differences between patients. Regions of interest are usually small and the retrieval system needs to focus on this region. Results show that GIFT is capable of supplying similar cases, particularly when using relevance feedback, on such a large database. Usual image retrieval is based on a single image as search target but for this application we have to select images by case as similar cases need to be found and not images. A few false positive cases often remain in the results but they can be sorted out quickly by the surgeons. Image retrieval can well be used for the planning of operations by supplying similar cases. A variety of challenges has been identified and partly solved (varying luminosity, small region of interested, case-based instead of image-based). This article mainly presents a case study to identify potential benefits and problems. Several steps for improving the system have been identified as well and will be described at the end of the paper.

An automatic system for the person identification based on comparison of a camera image and a document photo presents several design dfficulties. The main complication arises from the fact that the document photo may represent quite a different appearance of the test person, i.e., an older version, sometimes by years, or with a different hairstyle, facial expression and lighting conditions. In addition the scanned photo is usually of a much lower quality. In this paper elements of a system for a face comparison between camera images and archival photo albums are discussed and experimental results are given. Such a system may be instrumental for document control or for police investigations.

... Samy Ait-Aoudia, Ramdane Mahiou, Elhachemi Guerrout ESI - Ecole nationale Supérieure en Informatique, BP 68M, O-Smar 16270 Algiers, Algeria s_ait_aoudia@esi.dz, r_mahiou@esi.dz, e_guerrout@esi.dz ... Potts model The Potts model is a generalization of the Ising model. ...

The application of data mining (DM) in healthcare is increasing. Healthcare organizations generate and collect large voluminous and heterogeneous information daily and DM helps to uncover some interesting patterns, which leads to the manual tasks elimination, easy data extraction directly from records, to save lives, to reduce the cost of medical services and to enable early detection of diseases. These patterns can help healthcare specialists to make forecasts, put diagnoses, and set treatments for patients in health facilities. This work overviews DM methods and main issues. Three case studies illustrate DM in healthcare applications: (i) In-Vitro Fertilization; (ii) Content-Based Image Retrieval (CBIR); and (iii) Organ transplantation.

A major challenge in biomedical Content-Based Image Retrieval (CBIR) is to achieve meaningful mappings that minimize the semantic gap between the high-level biomedical semantic concepts and the low-level visual features in images. This paper presents a comprehensive learning-based scheme toward meeting this challenge and improving retrieval quality. The article presents two algorithms: a learning-based feature selection and fusion algorithm and the Ranking Support Vector Machine (Ranking SVM) algorithm. The feature selection algorithm aims to select 'good' features and fuse them using different similarity measurements to provide a better representation of the high-level concepts with the low-level image features. Ranking SVM is applied to learn the retrieval rank function and associate the selected low-level features with query concepts, given the ground-truth ranking of the training samples. The proposed scheme addresses four major issues in CBIR to improve the retrieval accuracy: image feature extraction, selection and fusion, similarity measurements, the association of the low-level features with high-level concepts, and the generation of the rank function to support high-level semantic image retrieval. It models the relationship between semantic concepts and image features, and enables retrieval at the semantic level. We apply it to the problem of vertebra shape retrieval from a digitized spine x-ray image set collected by the second National Health and Nutrition Examination Survey (NHANES II). The experimental results show an improvement of up to 41.92% in the mean average precision (MAP) over conventional image similarity computation methods.

Image alignment refers to finding the best transformation from a fixed reference image to a new image of a scene. This process is often guided by similarity measures between images, computed based on the image data. However, in time-critical applications state-of-the-art methods for computing similarity are too slow. Instead of using all the image data to compute similarity, one can use a subset of pixels to improve the speed, but often this comes at the cost of reduced accuracy. This makes the problem of image alignment a natural application domain for deliberation control using anytime algorithms. However, almost no research has been done in this direction. In this paper, we present anytime versions for the computation of two common image similarity measures: mean squared difference and mutual information. Off-line, we learn a performance profile specific to each measure, which is then used on-line to select the appropriate amount of pixels to process at each optimization step. Wh...

Learning a measure of similarity between pairs of objects is a fundamental prob- lem in machine learning. It stands in the core of classificati on methods like kernel machines, and is particularly useful for applications like searching for images that are similar to a given image or finding videos that are rel evant to a given video. In these tasks,

Use of XML offers a structured approach for representing information while maintaining separation of form and content. XML information retrieval is different from standard text retrieval in two aspects: the XML structure may be of interest as part of the query; and the information does not have to be text. In this paper, we describe an investigation of approaches to retrieve text and images from a large collection of XML documents, performed in the course of our participation in the INEX 2006 Ad Hoc and Multimedia tracks. We evaluate three information retrieval similarity measures: Pivoted Cosine, Okapi BM25 and Dirichlet. We show that on the INEX 2006 Ad Hoc queries Okapi BM25 is the most effective among the three similarity measures used for retrieving text only, while Dirichlet is more suitable when retrieving heterogeneous (text and image) data.

Video contains multiple types of audio and visual information, which are difficult to extract, combine or trade-off in general video information retrieval. This paper provides an evaluation on the effects of different types of information used for video retrieval from a video collection. A number of different sources of information are present in most typical broadcast video collections and can be exploited for information retrieval. We will discuss the contributions of automatically recognized speech transcripts, image similarity matching, and video OCR in the contexts of experiments performed as part of 2001 TREC Video Retrieval Track evaluation performed by the National Institute of Standards and Technology. For the queries used in this evaluation, image matching and video OCR proved to be the most important aspects of video information retrieval.