Kwabena Agyepong

Gradient Vector Flow Field and Mass Region Extraction in Digital Mammograms

Mass detection is one of the main computer-aided mammographic breast cancer detection techniques. Precisely selecting the regions that contain masses is an important step in mass segmentation using mammographic computer-aided detection. In this paper, an algorithm for extracting mass regions in digital mammograms is proposed, in which we use adaptive histogram equalization to enhance mammograms, use a gradient vector flow field to generate region boundaries, select N candidate locations according to the means and the standard deviations of intensities of the points with top brightness, use these points and the region boundaries to generate the convex hulls of the regions as the mass regions. 161 down-sampled mammogram images from the Digital Database for Screening Mammography project were test, and a detection rate of 82.6% is obtained. The experimental results indicated that the method is efficient and robust.

An Image Enhancement Algorithm Based on a Contrast Measure in the Wavelet Domain for Screening Mammograms

Currently, radiologists mainly use their eyes to discern cancer when they screen the mammograms. However, in many cases, cancer is not easily detected by the eyes because of bad imaging conditions. In order to improve the diagnostic rate of cancer, image enhancement technology is often used to enhance the image and aid the radiologists. In this paper, we developed a new image enhancement technology in the wavelet domain for radiologists to screen mammograms. The new image enhancement algorithm has several advantages. First, the image enhancement is based on a contrast measure defined in the wavelet domain which matches the human vision system better. The enhanced images are therefore more suitable for the human eye; second, the image enhancement is carried on in the wavelet domain which saves time if the image is compressed by JPEG2000. The algorithm was tested by an expert and the results are progressive.

Multisensory data exploitation using advanced image fusion and adaptive colorization

Multisensory data usually present complimentary information such as visual-band imagery and infrared imagery. There is strong evidence that the fused multisensor imagery increases the reliability of interpretation, and the colorized multisensor imagery improves observer performance and reaction times. In this paper, we propose an optimized joint approach of image fusion and colorization in order to synthesize and enhance multisensor imagery such that the resulting imagery can be automatically analyzed by computers (for target recognition) and easily interpreted by human users (for visual analysis). The proposed joint approach provides two sets of synthesized images, a fused image in grayscale and a colorized image in color using a fusion procedure and a colorization procedure, respectively. The proposed image fusion procedure is based on the advanced discrete wavelet (aDWT) transform. The fused image quality (IQ) can be further optimized with respect to an IQ metric by implementing an iterative aDWT procedure. On the other hand, the daylight coloring technique renders the multisensor imagery with natural colors, which human users are use to observing in everyday life. We hereby propose to locally colorize the multisensor imagery segment by mapping the color statistics of the multisensor imagery to that of the daylight images, with which the colorized images resemble daylight pictures. This local coloring procedure also involves histogram analysis, image segmentation, and pattern recognition. The joint fusion and colorization approach can be performed automatically and adaptively regardless of the image contents. Experimental results with multisensor imagery showed that the fused image is informative and clear, and the colored image appears realistic and natural. We anticipate that this optimized joint approach for multisensor imagery will help improve target recognition and visual analysis.

Design of Nonlinear Frequency Controller for Isolated Thermal Power System with Generation Rate Constraint

A nonlinear control scheme for thermal power system frequency control is described. Nonlinear controller is developed through describing function stability analysis of the system Controller ensures stability of the system for any input. Simulation results are given showing the smooth operation of described control scheme. Paper describes in detail the stability analysis and controller design.

Component-based target recognition inspired by human vision

In contrast with machine vision, human can recognize an object from complex background with great flexibility. For example, given the task of finding and circling all cars (no further information) in a picture, you may build a virtual image in mind from the task (or target) description before looking at the picture. Specifically, the virtual car image may be composed of the key components such as driver cabin and wheels. In this paper, we propose a component-based target recognition method by simulating the human recognition process. The component templates (equivalent to the virtual image in mind) of the target (car) are manually decomposed from the target feature image. Meanwhile, the edges of the testing image can be extracted by using a difference of Gaussian (DOG) model that simulates the spatiotemporal response in visual process. A phase correlation matching algorithm is then applied to match the templates with the testing edge image. If all key component templates are matched with the examining object, then this object is recognized as the target. Besides the recognition accuracy, we will also investigate if this method works with part targets (half cars). In our experiments, several natural pictures taken on streets were used to test the proposed method. The preliminary results show that the component-based recognition method is very promising.

Design of Adaptive Neural Network Frequency Controller for Performance Improvement of an Isolated Thermal Power System

An adaptive neural network control scheme for thermal power system is described. Neural network control scheme does not require off-line training. The online tuning algorithm and neural network architecture are described and a stability proof is given. The performance of the controller is illustrated via simulation for different changes in process parameters and for different disturbances. Performance of neural network controller is compared with conventional proportional-integral control scheme for frequency control in thermal power systems.

A study of mammographic mass retrieval based on shape and texture descriptors

Content-based mass image retrieval technology, utilizing both shape and texture features, is investigated in this paper. In order to retrieve similar mass patterns that help improve clinical diagnosis, the performance of mass retrieval using curvature scale space descriptors (CSSDs) and R-transform descriptors was mainly studied. The mass contours in the DDSM database (Univ. of South Florida) were preprocessed to eliminate curl cases, which is very important for the extraction of features. The peak extraction method from a CSS contour map by circular shift and CSSDs matching method were introduced. Preliminary experiments show that the performance of CSSDs and R-transform descriptors outperform other features such as moment invariants, normalized Fourier descriptors (NFDs), and the combined texture feature. By combining CSSDs with R-transform descriptors and the texture features based on Gray-level Co-occurrence Matrices (GLCMs), the experiments show that the hybrid method gives a better performance in mass image retrieval than CSSDs or R-transform descriptors.

Gradient Vector Flow Fields and Spiculated Mass Detection in Digital Mammography Images

In this paper, we proposed an algorithm for spiculated mass detection using digital down-sampled mammography images. In the algorithm, two-resulotion data is generated with a wavelet transform; For each resolution data, two gradient vector flow fields features, along with the standard deviation of a local edge orientation histogram, the mean, the standard deviation, and the standard deviation of the folded gradient orientations are extracted; a neural network classifier is used to generate spiculated mass masks; and the masks are filtered based on local relative intensity of the mammography images. The algorithm was tested using 200 mammograms including 100 massive images and 100 normal images from DDSM [17], in which FPI/TP of 1.0/0.88 and area of 0.71 under the ROC curve were obtained. The experimental results showed that the proposed method is efficient and robust.