S. A. R. Abu-Bakar

Watermarking techniques used in medical images: a survey.

The ever-growing numbers of medical digital images and the need to share them among specialists and hospitals for better and more accurate diagnosis require that patients’ privacy be protected. As a result of this, there is a need for medical image watermarking (MIW). However, MIW needs to be performed with special care for two reasons. Firstly, the watermarking procedure cannot compromise the quality of the image. Secondly, confidential patient information embedded within the image should be flawlessly retrievable without risk of error after image decompressing. Despite extensive research undertaken in this area, there is still no method available to fulfill all the requirements of MIW. This paper aims to provide a useful survey on watermarking and offer a clear perspective for interested researchers by analyzing the strengths and weaknesses of different existing methods.

A discrimination method for the detection of pneumonia using chest radiograph.

This paper presents a statistical method for the detection of lobar pneumonia when using digitized chest X-ray films. Each region of interest was represented by a vector of wavelet texture measures which is then multiplied by the orthogonal matrix Q(2). The first two elements of the transformed vectors were shown to have a bivariate normal distribution. Misclassification probabilities were estimated using probability ellipsoids and discriminant functions. The result of this study recommends the detection of pneumonia by constructing probability ellipsoids or discriminant function using maximum energy and maximum column sum energy texture measures where misclassification probabilities were less than 0.15.

A Heuristic Automatic and Robust ROI Detection Method for Medical Image Warermarking

This paper presents an automatic region of interest (ROI) segmentation method for application of watermarking in medical images. The advantage of using this scheme is that the proposed method is robust against different attacks such as median, Wiener, Gaussian, and sharpening filters. In other words, this technique can produce the same result for the ROI before and after these attacks. The proposed algorithm consists of three main parts; suggesting an automatic ROI detection system, evaluating the robustness of the proposed system against numerous attacks, and finally recommending an enhancement part to increase the strength of the composed system against different attacks. Results obtained from the proposed method demonstrated the promising performance of the method.

Noise reduction in iris recognition using multiple thresholding

Iris recognition is known to be as one of the most accurate biometric modalities. In iris image processing, the unwanted textures in the iris region such as those belong to pupil, eyelashes, eyelids, shadows and light reflections are defined as noises. These unwanted noises have strong gray values which cause wrong threshold value selection and thus, result in reducing the performance of the iris recognition system. In this paper, we proposed a multiple thresholding method for detection of eyelids, eyelash textures and light reflections and pupil pixels. The threshold values related to these noises are selected based on the information obtained from the histogram of the normalized iris image. The proposed method was applied to the CASIA V.3 iris image database, version three, from the institute of automation, Chinese academy of science and has 99.62% recognition rate with 0.04 false rejection rate (FRR).

A statistical interpretation of the chest radiograph for the detection of pulmonary tuberculosis

This paper presents a statistical interpretation of the chest radiograph for the detection of pulmonary tuberculosis (PTB). Each region of interest was represented by a vector of wavelet texture measures which is then multiplied by the orthogonal matrix Q. The first two elements of the transformed vectors were shown to have a bivariate normal distribution. Misclassification probabilities were estimated using probability ellipsoids and discriminant functions. The most important result of this study recommends the detection of pulmonary tuberculosis by constructing discriminant function using maximum column sum energy texture measures where the misclassification probabilities were less than 0.15. In the validation exercise, the proposed discriminant procedure yielded 94% correct classification rate.

A robust medical image watermarking against salt and pepper noise for brain MRI images

The ever-growing numbers of medical digital images and the need to share them among specialists and hospitals for better and more accurate diagnosis require that patients’ privacy be protected. During the transmission of medical images between hospitals or specialists through the network, the main priority is to protect a patient’s documents against any act of tampering by unauthorised individuals. Because of this, there is a need for medical image authentication scheme to enable proper diagnosis on patient. In addition, medical images are also susceptible to salt and pepper impulse noise through the transmission in communication channels. This noise may also be intentionally used by the invaders to corrupt the embedded watermarks inside the medical images. A common drawback of existing watermarking methods is their weakness against salt and pepper noise. The research carried out in this work addresses the issue of designing a new watermarking method that can withstand high density of salt and pepper noise for brain MRI images. For this purpose, combination of a spatial domain watermarking method, channel coding and noise filtering schemes are used. The region of non-interest (RONI) of MRI images from five different databases are used as embedding area and electronic patient record (EPR) is considered as embedded data. The quality of watermarked image is evaluated using Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index (SSIM), and the accuracy of the extracted watermark is assessed in terms of Bit Error Rate (BER).

Crowd counting using statistical features based on curvelet frame change detection

Automatic counting for moving crowds in digital images is an important application in computer artificial intelligence, especially for safety and management purposes. This paper presents a new method to estimate the size of a crowd. The new algorithm depends on sequential frame differences to estimate the crowd size in a scene. However, relying only on these simple differences adds more constraints for extracting sufficient crowd descriptors. A curvelet transform is employed to achieve that goal. Every two sequential frames are transformed into multi-resolution and multi-direction formats, and then the frame differences are detected at every subband in the curvelet domain. Statistical features out of each subband are then calculated, and the collected features from all subbands are considered as a descriptor vector for the crowd in the scene. Finally, a neural network is manipulated to map the descriptor vectors into predicted counts. The experimental results show that the proposed curvelet statistical features are more robust and provide crowd counting with higher accuracy than previous approaches.

Iris code matching using adaptive Hamming distance

The most popular metric distance used in iris code matching is Hamming distance. In this paper, we improve the performance of iris code matching stage by applying adaptive Hamming distance. Proposed method works with Hamming subsets with adaptive length. Based on density of masked bits in the Hamming subset, each subset is able to expand and adjoin to the right or left neighbouring bits. The adaptive behaviour of Hamming subsets increases the accuracy of Hamming distance computation and improves the performance of iris code matching. Results of applying proposed method on Chinese Academy of Science Institute of Automation, CASIA V3.3 shows performance of 99.96% and false rejection rate 0.06.

Prediction of soluble solid content of starfruit using spectral imaging combined with partial least squares and support vector regression

Spectral imaging technique such as hyperspectral and multispectral imaging is a combination of imaging and spectroscopy. This powerful technique can provide samples of spectral images, which can be used to analyze a number of fruit properties. The aim of this study is to develop calibration or predictive model for determining soluble solid content (SSC) of starfruit samples based on their spectral images. Partial least squares (PLSR) and support vector regression (SVR) techniques were applied to build the relationship between the mean spectral data and the reference value. The mean spectral data was extracted from spectral images of each starfruit samples. The simple template for region of interest (ROI) selection and five optimal wavelengths (565.2, 677.2, 736, 873.2 and 943.2 nm) as proposed in previous study were used for extraction of the mean spectral data. The result showed that the calibration model with PLSR and SVR had better performance than the previous study. Moreover, the calibration model with SVR was the best performance for prediction of SSC value of starfruit.

A hybrid skin color detection using HSV and YCgCr color space for face detection

Skin detection is a key aspect of many computer vision applications including face detection, person identification, illicit content detection and other related applications. In this paper, a skin detection method is proposed combining two color spaces HSV (Hue, Saturation, Value) and YCgCr (luminance, chrominance in green, chrominance in red). The S, Cg and Cr components are used to form a hybrid SCgCr color space. Skin detection results show that, the proposed method can respond well to different skin color tones with less sensitivity to skin-like background pixels. It is also shown that higher face detection rate can be achieved when applied to face detection problem.