Kresimir Delac

SCface --- surveillance cameras face database

In this paper we describe a database of static images of human faces. Images were taken in uncontrolled indoor environment using five video surveillance cameras of various qualities. Database contains 4,160 static images (in visible and infrared spectrum) of 130 subjects. Images from different quality cameras should mimic real-world conditions and enable robust face recognition algorithms testing, emphasizing different law enforcement and surveillance use case scenarios. In addition to database description, this paper also elaborates on possible uses of the database and proposes a testing protocol. A baseline Principal Component Analysis (PCA) face recognition algorithm was tested following the proposed protocol. Other researchers can use these test results as a control algorithm performance score when testing their own algorithms on this dataset. Database is available to research community through the procedure described at www.scface.org.

Independent Comparative Study of PCA, ICA, and LDA on the FERET Data Set

Face recognition is one of the most successful applications of image analysis and understanding and has gained much attention in recent years. Various algorithms were proposed and research groups across the world reported different and often contradictory results when comparing them. The aim of this paper is to present an independent, comparative study of three most popular appearance‐based face recognition projection methods (PCA, ICA, and LDA) in completely equal working conditions regarding preprocessing and algorithm implementation. We are motivated by the lack of direct and detailed independent comparisons of all possible algorithm implementations (e.g., all projection–metric combinations) in available literature. For consistency with other studies, FERET data set is used with its standard tests (gallery and probe sets). Our results show that no particular projection–metric combination is the best across all standard FERET tests and the choice of appropriate projection–metric combination can only be made for a specific task. Our results are compared to other available studies and some discrepancies are pointed out. As an additional contribution, we also introduce our new idea of hypothesis testing across all ranks when comparing performance results.

A Survey of Image Processing Algorithms in Digital Mammography

Mammography is at present the best available technique for early detection of breast cancer. The most common breast abnormalities that may indicate breast cancer are masses and calcifications. In some cases, subtle signs that can also lead to a breast cancer diagnosis, such as architectural distortion and bilateral asymmetry, are present. Breast abnormalities are defined with wide range of features and may be easily missed or misinterpreted by radiologists while reading large amount of mammographic images provided in screening programs. To help radiologists provide an accurate diagnosis, a computer-aided detection (CADe) and computer-aided diagnosis (CADx) algorithms are being developed. CADe and CADx algorithms help reducing the number of false positives and they assist radiologists in deciding between follow up and biopsy. This chapter gives a survey of image processing algorithms that have been developed for detection of masses and calcifications. An overview of algorithms in each step (segmentation step, feature extraction step, feature selection step, classification step) of the mass detection algorithms is given. Wavelet detection methods and other recently proposed methods for calcification detection are presented. An overview of contrast enhancement and noise equalization methods is given as well as an overview of calcification classification algorithms.

Recent Advances in Face Recognition

ing and non-profit use of the material is permitted with credit to the source. Statements and opinions expressed in the chapters are these of the individual contributors and not necessarily those of the editors or publisher. No responsibility is accepted for the accuracy of information contained in the published articles. Publisher assumes no responsibility liability for any damage or injury to persons or property arising out of the use of any materials, instructions, methods or ideas contained inside. After this work has been published by the In-Teh, authors have the right to republish it, in whole or part, in any publication of which they are an author or editor, and the make other personal use of the work.

Appearance-based statistical methods for face recognition

Different statistical methods for face recognition have been proposed in recent years. They mostly differ in the type of projection and distance measure used. The aim of this paper is to give an overview of most popular statistical subspace methods for face recognition task. Theoretical aspects of three algorithms will be considered and some reported performance evaluations will be given.

Face recognition in JPEG and JPEG2000 compressed domain

In this paper we investigate the potential of performing face recognition in JPEG and JPEG2000 compressed domain. This is achieved by avoiding full decompression and using transform coefficients as input to face recognition algorithms. We propose a new comparison methodology and by employing it show that face recognition can efficiently be implemented directly into compressed domain. In the first part of our experiment we use all the available transform coefficients and show that recognition rates are comparable and in some cases even higher than recognition rates obtained by using pixels from uncompressed images (standard face recognition approach). In the second part, we propose an effective coefficient preselection method (applicable both in JPEG and JPEG2000 compressed domain). Our results show that by using the proposed method, recognition rates can be significantly improved while additionally reducing computational time. Finally, we propose what a hypothetical compressed domain face recognition system should look like.

Breast Density Classification Using Multiple Feature Selection

Mammography as an x-ray method usually gives good results for lower density breasts while higher breast tissue densities significantly reduce the overall detection sensitivity and can lead to false negative results. In automatic detection algorithms knowledge about breast density can be useful for setting an appropriate decision threshold in order to produce more accurate detection. Because the overall intensity of mammograms is not directly correlated with the breast density we have decided to observe breast density as a texture classification problem. In this paper we propose breast density classification using feature selection process for different classifiers based on grayscale features of first and second order. In feature selection process different selection methods were used and obtained results show the improvement on overall classification by choosing the appropriate method and classifier. The classification accuracy has been tested on the mini-MIAS database and KBD-FER digital mammography database with different number of categories for each database. Obtained accuracy stretches between 97.2 % and 76.4 % for different number of categories.

Effects of JPEG and JPEG2000 compression on face recognition

In this paper we analyse the effects that JPEG and JPEG2000 compression have on subspace appearance-based face recognition algorithms. This is the first comprehensive study of standard JPEG2000 compression effects on face recognition, as well as an extension of existing experiments for JPEG compression. A wide range of bitrates (compression ratios) was used on probe images and results are reported for 12 different subspace face recognition algorithms. Effects of image compression on recognition performance are of interest in applications where image storage space and image transmission time are of critical importance. It will be shown that not only that compression does not deteriorate performance but it, in some cases, even improves it slightly. Some unexpected effects will be presented (like the ability of JPEG2000 to capture the information essential for recognizing changes caused by images taken later in time) and lines of further research suggested.

Image Compression Effects in Face Recognition Systems

With the growing number of face recognition applications in everyday life, image- and video-based recognition methods are becoming important research topic (Zhao et al., 2003). Effects of pose, illumination and expression are issues currently most studied in face recognition. So far, very little has been done to investigate the effects of compression on face recognition, even though the images are mainly stored and/or transported in a compressed format. Still-to-still image experimental setups are often researched, but only in uncompressed image formats. Still-to-video research (Zhou et al., 2003) mostly deals with issues of tracking and recognizing faces in a sense that still uncompressed images are used as a gallery and compressed video segments as probes. In this chapter we analyze the effects that standard image compression methods - JPEG (Wallace, 1991) and JPEG2000 (Skodras et al., 2001) - have on two well known subspace appearance-based face recognition algorithms: Principal Component Analysis - PCA (Turk & Pentland, 1991), Linear Discriminant Analysis - LDA (Belhumeur et al., 1996) and Independent Component Analysis - ICA (Bartlett et al., 2002). We use McNemars hypothesis test (Beveridge et al., 2001 ; Delac et al., 2006) when comparing recognition accuracy in order to determine if the observed outcomes of the experiments are statistically important or a matter of chance. Following the idea of a reproducible research, a comprehensive description of our experimental setup is given, along with details on the choice of images used in the training and testing stage, exact preprocessing steps and recognition algorithms parameters setup. Image database chosen for the experiments is the grayscale portion of the FERET database (Phillips et al., 2000) and its accompanying protocol for face identification, including standard image gallery and probe sets. Image compression is performed using standard JPEG and JPEG2000 coder implementations and all experiments are done in pixel domain (i.e. the images are compressed to a certain number of bits per pixel and then uncompressed prior to use in recognition experiments). The recognition systems overall setup we test is twofold. In the first part, only probe images are compressed and training and gallery images are uncompressed (Delac et al., 2005). This setup mimics the expected first step in implementing compression in real-life face recognition applications: an image captured by a surveillance camera is probed to an existing high-quality gallery image. In the second part, a leap towards justifying fully compressed domain face recognition is taken by using compressed images in both training and testing stage (Delac, 2006). We will show that, contrary to common opinion, compression does not deteriorate performance but it even improves it slightly in some cases. We will also suggest some prospective lines of further research based on our findings.

Statistics in face recognition: analyzing probability distributions of PCA, ICA and LDA performance results

In this paper we address the issue of evaluating face recognition algorithms using descriptive statistical tools. By using permutation methodology in a Monte Carlo sampling procedure, we investigate recognition rate results probability distributions of some well-known algorithms (namely, PCA, ICA and LDA). With a lot of contradictory literature on comparisons of those algorithms, we believe that this kind of independent study is important and serves to better understanding of each algorithm. We show how simplistic approach to comparing these algorithms can be misleading and propose a full statistical methodology to be used in future reports. By reporting detailed descriptive statistical results, this paper is the only available detailed report on PCA, ICA and LDA comparative performance currently available in literature. Our experiments show that the exact choice of images to be in a gallery or in a probe set has great effect on recognition results and this fact further emphasizes the importance of reporting detailed results. We hope that this study helps to advance the state of experiment design in computer vision.