Image Repositories Research Articles

ShadowPlay2.5D

An immersive experience brought about by virtual reality can potentially enhance the appreciation of classical Chinese poetry, which is difficult to describe clearly in everyday language or ordinary media. However, making 3-dimensional illustrations for a 360-degree display in virtual reality is usually a labor-intensive and time-consuming procedure and hard to master for non-professional media creators, such as teachers. Motivated by the homology theory of classical Chinese poetry and painting, we propose an image-based approach of building 2.5-dimensional immersive stories to visualize classical Chinese poetry. Specifically, using Chinese shadow play as a metaphor, we have designed and implemented ShadowPlay2.5D , a sketch-based authoring tool to help novices create 360-degree videos of classical Chinese poetry easily. To ensure coverage of the diverse themes in Chinese poetry and preserve the sense of culture, we build a Chinese ink-painting style image repository of essential poetic elements identified via crowdsourcing. To facilitate construction of 2.5-dimensional scenes, we design features that support puppet-like animation, instancing, and camera organization in a 3-dimensional environment. Through two user studies, we show that ShadowPlay2.5D can help novices make a short 360-degree video in about 10--15 minutes, and the 2.5D stylized illustrations created can bring about a better immersive experience for poetry appreciation.

Deep learning model with low-dimensional random projection for large-scale image search

Developing deep learning models that can scale to large image repositories is increasingly gaining significant efforts in the domain of image search. The current deep neural networks rely on the computational power of accelerators (e.g. GPUs) to tackle the processing limitations associated with features extraction and model training. This paper introduces and investigates a deep model of Convolutional Neural Networks (CNNs) to efficiently extract, index, and retrieve images in the context of large-scale Content-Based Image Retrieval (CBIR). Random Maclaurin projection is used to generate low-dimensional image descriptors and their discriminating efficiency is evaluated on standard image datasets. The scalability of deep architectures is also evaluated on one million image dataset over a High-Performance Computing (HPC) platform, which is assessed in terms of the retrieval accuracy, speed of features extraction and memory costs. Additionally, the controlling GPU kernels of the proposed model are examined under several optimization factors to evaluate their impact on the processing and retrieval performance. The experimental results show the effectiveness of the proposed model in the retrieval accuracy, GPU utilisation, speed of features extraction, and storage of image indexing.

Establishment of a Nationwide Korean Imaging Cohort of Coronavirus Disease 2019.

BackgroundThe Korean Society of Thoracic Radiology (KSTR) recently constructed a nation-wide coronavirus disease 2019 (COVID-19) database and imaging repository, referred to the Korean imaging cohort of COVID-19 (KICC-19) based on the collaborative efforts of its members. The purpose of this study was to provide a summary of the clinico-epidemiological data and imaging data of the KICC-19.MethodsThe KSTR members at 17 COVID-19 referral centers retrospectively collected imaging data and clinical information of consecutive patients with reverse transcription polymerase chain reaction-proven COVID-19 in respiratory specimens from February 2020 through May 2020 who underwent diagnostic chest computed tomography (CT) or radiograph in each participating hospital.ResultsThe cohort consisted of 239 men and 283 women (mean age, 52.3 years; age range, 11–97 years). Of the 522 subjects, 201 (38.5%) had an underlying disease. The most common symptoms were fever (n = 292) and cough (n = 245). The 151 patients (28.9%) had lymphocytopenia, 86 had (16.5%) thrombocytopenia, and 227 patients (43.5%) had an elevated CRP at admission. The 121 (23.4%) needed nasal oxygen therapy or mechanical ventilation (n = 38; 7.3%), and 49 patients (9.4%) were admitted to an intensive care unit. Although most patients had cured, 21 patients (4.0%) died. The 465 (89.1%) subjects underwent a low to standard-dose chest CT scan at least once during hospitalization, resulting in a total of 658 CT scans. The 497 subjects (95.2%) underwent chest radiography at least once during hospitalization, which resulted in a total of 1,475 chest radiographs.ConclusionThe KICC-19 was successfully established and comprised of 658 CT scans and 1,475 chest radiographs of 522 hospitalized Korean COVID-19 patients. The KICC-19 will provide a more comprehensive understanding of the clinical, epidemiological, and radiologic characteristics of patients with COVID-19.

Implementing Relevance Feedback for Content-Based Medical Image Retrieval

Content-based image medical retrieval (CBMIR) is a technique for retrieving medical images on the basis of automatically derived image features such as colour, texture and shape. There are many applications of CBMIR, such as teaching, research, diagnosis and electronic patient records. The retrieval performance of a CBMIR system depends mainly on the representation of image features, which researchers have studied extensively for decades. Although a number of methods and approaches have been suggested, it remains one of the most challenging problems in current (CBMIR) studies, largely due to the well-known “semantic gap” issue that exists between machine-captured low-level image features and human-perceived high-level semantic concepts. There have been many techniques proposed to bridge this gap. This study proposes a novel relevance feedback retrieval method (RFRM) for CBMIR. The feedback implemented here is based on voting values performed by each class in the image repository. Here, eighteen using colour moments and GLCM texture features were extracted to represent each image and eight common similarity coefficients were used as similarity measures. After briefly researching using a single random image query, the top images retrieved from each class are used as voters to select the most effective similarity coefficient that will be used for the final searching process. Our proposed method is implemented on the Kvasir dataset, which has 4,000 images divided into eight classes and was recently widely used for gastrointestinal disease detection. Intensive statistical analysis of the results shows that our proposed RFRM method has the best performance for enhancing both recall and precision when it uses any group of similarity coefficients.

An Efficient and Effective Model Based on Mean Positive Examples for Social Image Annotation

Nowadays, with the rapid growth of imaging and social network, huge volumes of image data are produced and shared on social media. Social image annotation has been an important and challenging task in the fields of computer vision and machine learning, which can facilitate large-scale image retrieval, indexing, and management. The four most challenges of social image annotation are semantic gap, tag refinement, label-imbalance, and annotation efficiency. To address these issues, we propose an efficient and effective annotation method based on the Mean of Positive Examples (MPE) corresponding to each label. First, we refine user-provided noisy tags by our proposed local smoothing process, and consider the refined tags as key features in contrast to the previous methods that consider them as side information, which significantly improves annotation performance. Second, we propose a weighted trans-media similarity measure method that fuses all modality information in identifying proper neighbors, which promotes the semantic level and eases image annotation. Third, our MPE model gives equal importance to all labels, thus, improving the annotation performance of infrequent labels without sacrificing that of frequent labels. Fourth, our MPE model can dramatically decrease space-time overheads, since the time cost of annotating an image is unaffected by the size of the training image dataset, but relying on the size of label vocabulary. Therefore, our proposed method can be applied to real-world large-scale online social image repositories. Extensive experiments on both benchmark datasets demonstrate the effectiveness and efficiency of our MPE model.

DIVDS: Docker Image Vulnerability Diagnostic System

Since the development of Docker in 2013, container utilization projects have emerged in various fields. Docker has the advantage of being able to quickly share application build environments among developers through container technology, but it does not provide security guarantees for known security vulnerabilities inside Docker images. Since the Docker images are shared without a means of security vulnerability diagnostic, polluted Docker images can be distributed so that the Docker-based application build environments can be easily collapsed. In this paper, we introduce a Docker Image Vulnerability Diagnostic System (DIVDS) for a reliable Docker environment. The proposed DIVDS diagnoses Docker images when uploading or downloading the Docker images from a Docker image repository.

A Granular Level Feature Extraction Approach to Construct HR Image for Forensic Biometrics Using Small Training DataSet

In forensic biometrics old x-ray images are often used for identification and verification. The number of homicide cases has increased manyfold along with population growth. Forensic odontology is a less expensive method to solve these cases. In forensic odontology, the old dental x-ray images of a victim are used as ante mortem data to compare the pattern for identification. The success of this post mortem biometrics process totally depends on the brain perception of forensic odontologist. Low resolution (LR) image may create human brain perception error at the time of making decision both in case of disease diagnosis and forensic biometrics. In such context, a software solution could help to reconstruct high resolution (HR) image from LR medical x-ray image. However, methods like convolution neural network (CNN) require high volume of training images to reconstruct HR images. Unfortunately, the available medical x-ray image repository does not offer a large volume of training dataset. This work aims to overcome this data related issue and presents a granular level feature based HR grayscale medical x-ray image reconstruction mechanism from LR image. This method uses machine learning for HR image reconstruction. The proposed granular level feature extraction method generates adequate amount of training set from limited amount of training image sets. These granular level features contain the influence of neighboring points separately and its direction. These features are highly immune to noises and preserve important properties like edges. Polynomial regression model is used for HR value reconstruction. This image reconstruction provides satisfactory results for image database like dental radiographs datasets and the average structural similarity index (SSIM) metric reaches 0.9326.

Online Brain Image Repositories for Brain Disease Detection

Brain image analysis is an emerging area of researchers to improve the diagnosis process more fast and accurate. One of the difficulties is getting the clinical dataset of patients from hospitals to test the performance of the proposed methods. Therefore, numerous online brain image repositories are available to promote the research works. It has manually segmented results to evaluate the accuracy of the developed methods. Each repository has different file format and focused on different problems like skull stripping, tumorous image classification, tumor type categorization, tissue segmentation and tumor with substructure segmentation. This paper gives detail information on famous brain datasets with their purpose.

Generalized Multiscale RBF Networks and the DCT for Breast Cancer Detection

The use of the multiscale generalized radial basis function (MSRBF) neural networks for image feature extraction and medical image analysis and classification is proposed for the first time in this work. The MSRBF networks hold a simple and flexible architecture that has been successfully used in forecasting and model structure detection of input-output nonlinear systems. In this work instead, MSRBF networks are part of an integrated computer-aided diagnosis (CAD) framework for breast cancer detection, which holds three stages: an input-output model is obtained from the image, followed by a high-level image feature extraction from the model and a classification module aimed at predicting breast cancer. In the first stage, the image data is rendered into a multiple-input-single-output (MISO) system. In order to improve the characterisation, the nonlinear autoregressive with exogenous inputs (NARX) model is introduced to rearrange the available input-output data in a nonlinear way. The forward regression orthogonal least squares (FROLS) algorithm is then used to take advantage of the previous arrangement by solving the system as a model structure detection problem and finding the output layer weights of the NARX-MSRBF network. In the second stage, once the network model is available, the feature extraction takes place by stimulating the input to produce output signals to be compressed by the discrete cosine transform (DCT). In the third stage, we leverage the extracted features by using a clustering algorithm for classification to integrate a CAD system for breast cancer detection. To test the method performance, three different and well-known public image repositories were used: the mini-MIAS and the MMSD for mammography, and the BreaKHis for histopathology images. A comparison exercise was also made between different database partitions to understand the mammogram breast density effect in the performance since there are few remarks in the literature on this factor. Classification results show that the new CAD method reached an accuracy of 93.5% in mini-Mammo graphic image analysis society (mini-MIAS), 93.99% in digital database for screening mammography (DDSM) and 86.7% in the BreaKHis. We found that the MSRBF networks are able to build tailored and precise image models and, combined with the DCT, to extract high-quality features from both black and white and coloured images.

MULTI-TEMPORAL ANALYSIS OF DENSE AND SPARSE FORESTS’ RADAR BACKSCATTER USING SENTINEL-1A COLLECTION IN GOOGLE EARTH ENGINE

Abstract. Radar data has been historically expensive and complex to process. However, in this milieu of cloud-computing platforms and open-source datasets, radar data analysis has become convenient and can now be performed for more exploratory researches. This study aims to perform multi-temporal analysis of radar backscatter to characterize dense and sparse forest from Sentinel-1 images. The area of study are reforested sites under the National Greening Program (NGP) of the Philippines. Ground data were collected: (1) in 2019, from a 1.35 ha -site in Brgy. Calula, Ipil, Zamboanga Sibugay, (2) in 2019, from a 1.10 ha- site in Brgy. Cabatuanan, Basay, Negros Oriental, and (3) from PhilLiDAR 2 – Project 3: FRExLS’ 2.4 ha -validated site in Ubay, Bohol. SAR intensity values were derived from Sentinel-1 from Google Earth Engine, which is a cloud-based platform with a repository of satellite images and functionalities for data extraction and processing. The temporal variation in C-band radar backscatter from 2014 to 2018 were analyzed. The results show, for the whole period of analysis, that: in VH polarization, dense forest samples backscatter range from −11 to −18 dB in VH and −2 to -13 dB in VV; sparse forest samples range from −12 to -21 dB in VH and −7 to −14 dB in VV; ground samples range from −12 to −24 dB in VH and −6 to −15 dB in VV; and water samples range from −21 to −30 dB in VH and −11 to −26 dB in VV. Forest backscatter are expected to saturate over time, especially in dense forests. These variations are due to differences in forest species, landscape, environmental and climatic drivers, and phenomenon or interventions on the site.

Abstract C62: The Alliance of Ppancreatic Cancer Consortia for Biomarkers for Early Detection

Abstract Pancreatic cancer is the fourth leading cause of cancer death in the United States and the 5-year relative survival for patients diagnosed with pancreatic cancer is less than 10%. Early intervention is the key to a better survival outcome. Currently, there are no biomarkers that can reliably detect pancreatic cancer at an early stage or identify precursors that are destined to progress to malignancy. The Alliance of Pancreatic Cancer Consortia for Biomarkers for Early Detection was formed under the auspices of a workshop organized by the National Cancer Institute in partnership with the Kenner Family Research Fund and the Pancreatic Cancer Action Network (December 5, 2016). The second, biannual workshop of the Alliance was convened December 12, 2018. The organizers of this workshop included Drs. Ralph Hruban, Kenneth Kinzler, Stephen Pandol, Brian Wolpin, Walter Park, Samir Hanash, and Anirban Maitra. Investigators from four NCI-supported consortia on pancreatic cancer detection, including the Pancreatic Cancer Detection Consortium (PCDC), the Early Detection Research Network (EDRN), the Consortium for the Study of Chronic Pancreatitis, Diabetes and Pancreatic Cancer (CPDPC), and the Consortium for Molecular and Cellular Characterization of Screen-Detected Lesions (MCL), were invited to participate in the workshop. Other invited participants included members of the Kenner Family Research Fund, the Pancreatic Cancer Action Network, investigators from industry, and biostatisticians from NCI‘s Division of Cancer Prevention, Biometry Research Group, and from the EDRN. The keynote speakers included Dr. Ralph Hruban from Johns Hopkins University who spoke about the challenges in pancreatic cancer research, and Dr. James Abbruzzese from Duke Cancer Institute who provided an updated report from NCI’s Scientific Framework for Pancreatic Ductal Adenocarcinoma (PDAC), which met shortly before the workshop. The focus of the workshop included discussion of the molecular characterization of PDAC and its precursors and methodologies for early detection of pancreatic cancer, new imaging applications and artificial intelligence for mining big data, development and utilization of new methods for biomarker validation, and biospecimen science and resources. The members of the workshop agreed that the Alliance should immediately begin working on the development of a repository for images collected prior to the diagnosis of pancreatic cancer. These images could be shared with the Alliance members and would be useful for development of imaging biomarkers. Working with the NCI, the Alliance has begun establishing the common data elements (CDE) needed to develop a repository of pancreatic cancer images. The Alliance of Pancreatic Cancer Consortia provides a valuable forum for investigators and collaborators to discuss and address early detection of pancreatic cancer. Citation Format: Matthew R. Young, Natalie Abrams, Sharmistha Ghosh-Janjigian, Guillermo Marquez, Jo Ann Rinaudo, Sudhir Srivastava. The Alliance of Ppancreatic Cancer Consortia for Biomarkers for Early Detection [abstract]. In: Proceedings of the AACR Special Conference on Pancreatic Cancer: Advances in Science and Clinical Care; 2019 Sept 6-9; Boston, MA. Philadelphia (PA): AACR; Cancer Res 2019;79(24 Suppl):Abstract nr C62.

Call for Data Standardization: Lessons Learned and Recommendations in an Imaging Study.

Data sharing creates potential cost savings, supports data aggregation, and facilitates reproducibility to ensure quality research; however, data from heterogeneous systems require retrospective harmonization. This is a major hurdle for researchers who seek to leverage existing data. Efforts focused on strategies for data interoperability largely center around the use of standards but ignore the problems of competing standards and the value of existing data. Interoperability remains reliant on retrospective harmonization. Approaches to reduce this burden are needed. The Cancer Imaging Archive (TCIA) is an example of an imaging repository that accepts data from a diversity of sources. It contains medical images from investigators worldwide and substantial nonimage data. Digital Imaging and Communications in Medicine (DICOM) standards enable querying across images, but TCIA does not enforce other standards for describing nonimage supporting data, such as treatment details and patient outcomes. In this study, we used 9 TCIA lung and brain nonimage files containing 659 fields to explore retrospective harmonization for cross-study query and aggregation. It took 329.5 hours, or 2.3 months, extended over 6 months to identify 41 overlapping fields in 3 or more files and transform 31 of them. We used the Genomic Data Commons (GDC) data elements as the target standards for harmonization. We characterized the issues and have developed recommendations for reducing the burden of retrospective harmonization. Once we harmonized the data, we also developed a Web tool to easily explore harmonized collections. While prospective use of standards can support interoperability, there are issues that complicate this goal. Our work recognizes and reveals retrospective harmonization issues when trying to reuse existing data and recommends national infrastructure to address these issues.

Denoising scanner effects from multimodal MRI data using linked independent component analysis

Pooling magnetic resonance imaging (MRI) data across research studies, or utilizing shared data from imaging repositories, presents exceptional opportunities to advance and enhance reproducibility of neuroscience research. However, scanner confounds hinder pooling data collected on different scanners or across software and hardware upgrades on the same scanner, even when all acquisition protocols are harmonized. These confounds reduce power and can lead to spurious findings. Unfortunately, methods to address this problem are scant. In this study, we propose a novel denoising approach that implements a data-driven linked independent component analysis (LICA) to identify scanner-related effects for removal from multimodal MRI to denoise scanner effects. We utilized multi-study data to test our proposed method that were collected on a single 3T scanner, pre- and post-software and major hardware upgrades and using different acquisition parameters. Our proposed denoising method shows a greater reduction of scanner-related variance compared with standard GLM confound regression or ICA-based single-modality denoising. Although we did not test it here, for combining data across different scanners, LICA should prove even better at identifying scanner effects as between-scanner variability is generally much larger than within-scanner variability. Our method has great promise for denoising scanner effects in multi-study and in large-scale multi-site studies that may be confounded by scanner differences.

Neurocognitive markers of childhood abuse in individuals with PTSD: Findings from the INTRuST Clinical Consortium

To date, few studies have evaluated the contribution of early life experiences to neurocognitive abnormalities observed in posttraumatic stress disorder (PTSD). Childhood maltreatment is common among individuals with PTSD and is thought to catalyze stress-related biobehavioral changes that might impact both brain structure and function in adulthood. The current study examined differences in brain morphology (brain volume, cortical thickness) and neuropsychological performance in individuals with PTSD characterized by low or high self-reported childhood maltreatment, compared with healthy comparison participants. Data were drawn from the INjury and TRaUmatic STress (INTRuST) Clinical Consortium imaging repository, which contains MRI and self-report data for individuals classified as PTSD positive (with and without a history of mild traumatic brain injury [mTBI]), individuals with mTBI only, and healthy comparison participants. The final sample included 36 individuals with PTSD without childhood maltreatment exposure (PTSD, n = 30 with mTBI), 31 individuals with PTSD and childhood maltreatment exposure (PTSD + M, n = 26 with mTBI), and 114 healthy comparison participants without history of childhood maltreatment exposure (HC). The PTSD + M and PTSD groups demonstrated cortical thinning in prefrontal and occipital regions, and poorer verbal memory and processing speed compared to the HC group. PTSD + M participants demonstrated cortical thinning in frontal and cingulate regions, and poorer executive functioning relative to the PTSD and HC groups. Thus, neurocognitive features varied between individuals with PTSD who did versus did not have exposure to childhood maltreatment, highlighting the need to assess developmental history of maltreatment when examining biomarkers in PTSD.

Language-based colorization of scene sketches

Being natural, touchless, and fun-embracing, language-based inputs have been demonstrated effective for various tasks from image generation to literacy education for children. This paper for the first time presents a language-based system for interactive colorization of scene sketches, based on semantic comprehension. The proposed system is built upon deep neural networks trained on a large-scale repository of scene sketches and cartoonstyle color images with text descriptions. Given a scene sketch, our system allows users, via language-based instructions, to interactively localize and colorize specific foreground object instances to meet various colorization requirements in a progressive way. We demonstrate the effectiveness of our approach via comprehensive experimental results including alternative studies, comparison with the state-of-the-art methods, and generalization user studies. Given the unique characteristics of language-based inputs, we envision a combination of our interface with a traditional scribble-based interface for a practical multimodal colorization system, benefiting various applications. The dataset and source code can be found at https://github.com/SketchyScene/SketchySceneColorization.

The Enterprise Imaging Value Proposition.

As resources in the healthcare environment continue to wane, leaders are seeking ways to continue to provide quality care bounded by the constraints of a reduced budget. This manuscript synthesizes the experience from a number of institutions to provide the healthcare leadership with an understanding of the value of an enterprise imaging program. The value of such a program extends across the entire health system. It leads to operational efficiencies through infrastructure and application consolidation and the creation of focused support capabilities with increased depth of skill. An enterprise imaging program provides a centralized foundation for all phases of image management from every image-producing specialty. Through centralization, standardized image exchange functions can be provided to all image producers. Telehealth services can be more tightly integrated into the electronic medical record. Mobile platforms can be utilized for image viewing and sharing by patients and providers. Mobile tools can also be utilized for image upload directly into the centralized image repository. Governance and data standards are more easily distributed, setting the stage for artificial intelligence and data analytics. Increased exposure to all image producers provides opportunities for cybersecurity optimization and increased awareness.

The ICTBioMed NCIP Hub: Cancer research in a science gateway consortium

ICTBioMed, the International Consortium for Technology in Biomedicine is a consortium for biomedical collaborative research and the consortium partners believe in decentralization and democratization of science. The consortium works on strategies to develop ready-to-use research initiatives bridging the gap between research, implementation and ready-to-use science. The consortium has identified real science use cases with the help of the collaborative partners to get access of shared data and methods using the principles of science gateways. ICTBioMed has set up an NCIP Hub instance based upon the HUBzero framework and started using it for community research activities. It is an environment for scientific collaboration where researchers can access and take advantage of a variety of resources shared by others. The NCIP Hub supports members with shared access to data, tools, and standards across the cancer research community. Cancer researchers, in principle, have access to a wide range of data and software tools but in practice, however, they make limited use of these resources. Barriers that restrict access include the need to install software locally, inconsistent user interfaces, security concerns and poor documentation. ICTBioMed is developing a cancer science gateway that will greatly simplify analysis in cancer imaging research by offering a single point of entry and a unified interface for multiple tools, databases, and image repositories widely used by the community. The consortium aims at borderless research with a common goal to solve real life biomedical problems by facilitating collaborative research.

The BioDoseNet image repository used as a training tool for the dicentric assay

Purpose: This paper describes how the BioDoseNet image repository was used as a training tool for the dicentric assay.Materials and methods: The training was implemented in three phases: introduction to dicentric scoring, dose response curve elaboration and dose assessment exercise. Four labs without previous experience in the dicentric assay participated and four modules of the repository were used.Results: The labs become familiar with aberrations induced by ionizing radiation. The labs were able to generate data for the elaboration of a dose response curve and then successfully estimated doses and irradiated fractions in six blind samples.Conclusions: The performance of these laboratories during the exercise demonstrates the efficacy of the BioDoseNet image repository as a training tool and the utility of web based scoring for the dicentric assay community.

An Enhanced Framework for Content Based Medical Image Retrieval using Deep Neural Network

As the technology growth fuelled by low cost tech in the areas of compute, storage the need for faster retrieval and processing of data is becoming paramount for organizations. The medical domain predominantly for medical image processing with large size is critical for making life critical decisions. Healthcare community relies upon technologies for faster and accurate retrieval of images. Traditional, existing problem of efficient and similar medical image retrieval from huge image repository are reduced by Content Based Image Retrieval (CBIR) . The major challenging is an semantic gap in CBIR system among low and high level image features. This paper proposed, enhanced framework for content based medical image retrieval using DNN to overcome the semantic gap problem. It is outlines the steps which can be leveraged to search the historic medical image repository with the help of image features to retrieve closely relevant historic image for faster decision making from huge volume of database. The proposed system is assessed by inquisitive amount of images and the performance efficiency is calculated by precision and recall evaluation metrics. Experimental results obtained the retrieval accuracy is 79% based on precision and recall and this approach is preformed very effectively for image retrieval performance.

Performance Analysis of Distance Metric for Content Based Image Retrieval

Content based image retrieval uses different feature descriptors for image search and retrieval. For image retrieval from huge image repositories, the query image features are extracted and compares these features with the contents of feature repository. The most matching image is found and retrieved from the database. This mapping is done based on the distance calculated between feature vector of query image and the extracted feature vectors of images in the database. There are various distance measures used for comparing image feature vectors. This paper compares a set of distance measures using a set of features used for CBIR. The city-block distance measure gives the best results for CBIR.