Search

In this paper, optimization technique based on Ants Colony Algorithms (ACA) was studied for test node selection. The technique yields high faults detection and isolation rates with less number of nodes. This concept can also be applied in the domain of fault dictionary. In the technique proposed, the objective function consists of the fault detection rate defined as the number of faults detected divided by the total number of faults, and the fault isolation rate which is equal to the number of faults isolated divided by the number of fault detected. The transfer probability, also regarded as the rate of change of positions, is equal to the exponential power of - dij where dij is the difference between two points located in a two dimensional space. The updating procedure measures each solution against the last N solutions globally made by ACA. As soon as N solutions are available, their moving average X is computed; each new solution G is compared with X. If G value is lower than X value, the trail level of the last solution's move is increased, otherwise it's decreased. The efficiency of ACA-based technique was proved theoretically and by means of simulation. The source code proving the suitability of ACA was developed in Matlab. The both fault detection and isolation rates were compared with the rates generated by the AdvancedSCH, a platform developed by our research team. The comparison shows that ACA-based technique best suits for test node selection.

The combination of ordinary wavelet shrinkage with total variation minimization was successfully applied. In this paper, we apply the technique with respect to ridgelet coefficients. Firstly, a translation-invariant ridgelet transform is proposed. And then, an image denoising algorithm, based on ridgelet shrinkage and total variation minimization, is given. This algorithm preserves the important information of image and reduces the noise by thresholding small ridgelet coefficients. By replacing these thresholded coefficients by values minimizing the total variation, the algorithm reduces the pseudo-Gibbs artifacts. Experiment results show that this algorithm yields significantly superior image quality and higher peak signal to noise ratio (PSNR).

We propose a novel edge detection method based on fractal coding. We point out that the mean-square-error distance (MSED) of fractal coding can be used to extract edges of a fractal image effectively. The algorithm for edge detection based on fractal coding MSED is given and its improvement is discussed. Results indicate that since the new algorithm for edge detection avoids the gradient operation, it is insensitive to noise and suitable for object detection in natural scenes based on edges.

The watermarking technology is a new technique of embedding hidden images and data into an original image area. A watermarking method based on fractal self-similarity is given. Our watermarking method can be used to protect aptly important images and data. Hidden images and data can be retrieved directly from the image of hiding data. Experiments have shown that data or images are embedded in the whole image area and that the hidden images and data are difficult to be discovered and removed. Hence, the method can be used to protect effectively important images and data.