Wenbin Luo

An efficient detail-preserving approach for removing impulse noise in images

In this letter, we present a new efficient algorithm for the removal of impulse noise from corrupted images while preserving image details. The algorithm is based on the alpha-trimmed mean, which is a special case of the order-statistics filter. However, unlike some of the previous techniques involving order-statistics filters, the proposed method uses the alpha-trimmed mean only in impulse noise detection instead of pixel value estimation. Once a noisy pixel is identified, its value is replaced by a linear combination of its original value and the median of its local window. Extensive computer simulations indicate that our algorithm provides a significant improvement over many other existing techniques

Fast and robust watermarking of JPEG files

In this paper, we present a fast watermarking method that applies to JPEG images. JPEG is a standard image format supported by virtually all available image software applications. Our method manipulates the quantized DCT coefficients directly and can be implemented in real time. We use texture masking to embed a stronger watermark signal in certain texture areas. Watermark robustness to JPEG compression, additive Gaussian noise and image cropping attacks is studied with the proposed system. The relationship between watermark robustness and watermark position is also studied. A description of the method is included and results are presented along with a discussion of possible improvements.

Automated Analysis of Optic Nerve Images for Detection and Staging of Papilledema

PURPOSE To develop an automated system that analyzes digital fundus images for staging and monitoring of optic disc edema (i.e., papilledema), due to raised intracranial pressure. METHODS A total of 294 retrospective, digital photographs of the right and left eyes of 39 subjects with papilledema acquired over the span of 2 years were used. Software tools were developed to analyze three features of papilledema from digital fundus photographs: (1) sharpness of the optic disc border, (2) discontinuity along major vessels overlying the optic nerve, and (3) texture properties of the peripapillary retinal nerve fiber layer (RNFL). A classifier used these features to assign a grade of papilledema according to a standard protocol used by an expert neuro-ophthalmologist (RK). RESULTS The algorithm showed substantial agreement (κ = 0.71, P < 0.001) with the neuro-ophthalmologist when grading papilledema per patient. Vessel features showed statistical significance (P < 0.05) in differentiating grades 0, 1, and 2 from grades 3 and 4, whereas disc obscuration differentiated grades 0 or 1 from the rest (P < 0.05). CONCLUSIONS These results show that this algorithm can be used to automatically grade papilledema. The algorithm provides objective and quantitative assessment of the stage of papilledema with accuracy that is comparable to grading by a neuro-ophthalmologist. One application is in rapid assessment of digital optic nerve photographs acquired in clinical, intensive care, and emergency response settings by nonophthalmologists to evaluate for the presence and severity of papilledema, due to intracranial hypertension.

Impulse noise removal utilizing second-order difference analysis

The pending problem that research in random-valued impulse noise filtering has been facing is the inability to distinguish noisy values that do not occur as extreme outliers in comparison with other surrounding pixels. In this paper, we propose a new detection and filtering algorithm that consists of (1) a two-stage detection scheme that employs second-order difference between pixels to determine the integrity of the image pixels and (2) a noise filtering process that estimates the original value of each noisy pixel utilizing the information gathered from (1). Due to its unbiased detection criteria, this method treats both fixed-valued and random-valued noise with extremely high detection rate.

Automatic initialization of level set segmentation for application to optic disc margin identification

This paper presents a methodology for automatic initialization of a level-set segmentation algorithm to find the margin of the optic disc in fundus images as an indicator for glaucoma. Accurate segmentation of the optic disc is fundamental to the diagnosis and treatment of glaucoma. Our algorithm improves on previous applications by automatically finding the initialization points for ideal segmentation and more efficient convergence. Our algorithm was evaluated by comparing its output with images manually segmented by glaucoma specialists. Our results produced an average overlap with ground truth of 82%.

The modular design and implementation of an intelligent cruise control system

In this paper, we present the modular design and implementation of an intelligent cruise control system, following the principles of system of systems engineering. The proposed intelligent cruise control system can be seen as system of systems, which consists of several different subsystems. Different subsystems communicate with each other and existing electronic systems on the automobile via programmed microcontrollers. The proposed intelligent cruise control system aims to reduce the number of highway accidents and help eliminate road rage incidents. To do this, the implemented system slows down the car or steers away from the obstacle, when it is too late to slow down the car to a safe stop, to avoid collision. Using state-of-the-art algorithms and some inexpensive sensors around the vehicle, our proposed system can detect when it is safe to change lanes, or when to take over another car. Therefore, the proposed system can reduce the amount of attention required by the driver, reduce stress, and make driving a safer activity.

Simulation of animal behavior using neural networks

In this paper, we simulated some basic animal behavior by letting it “evolve” through generations. The animals, which are able to survive, will be given the privilege to breed the new generation and pass the good genes to the next generation. Our simulation environment consisted of a Prey-Predator-Food model. The prey and predators created in our simulation are objects that have sensors to make them aware of their environment, and they are able to move freely around the world. The food sources are stationary objects placed randomly throughout the world, which serve as targets for prey to search for and eat.

Impulse noise removal: Noise detection versus pixel estimation

This paper proposes a new finding regarding the different roles and significance of noise detection and pixel estimation utilized in the majority of impulse noise removing algorithms. Extensive experimental results will be presented to show that the importance of each method depends on the noise ratio. In other words, by particularly examining how the adaptive median filter (AMF) algorithm [18] employs the two aforementioned methods in comparison with their modified versions as well as the ideal-filtering algorithm, this paper will point out the role that each method plays. The significance of noise detection can be realized at low noise ratio while pixel estimation gains its efficiency as noise ratio increases.