Enhanced Genetic Algorithm Applied for Image Denoising Problem

Wavelet is one method that can be used as a step to recognition an individual . Wavelet is used to perform an image feature extraction . In the process of image texture analysis , showed the value of the coefficients of a wavelet . The magnitude of the coefficients obtained using wavelet feature extraction is influenced by several factors . Besides affected by the type of wavelet used , is also influenced by the magnitude of the wavelet decomposition level of itself. In the wavelet is known as the wavelet energy retention , which means the number of retained energy after undergoing a process of decomposition and cutting coefficients . During the decomposition process , the calculation for texture analysis is often a constraint . In order for the current calculation lighter texture analysis , necessary to the process of cutting coefficients based on the retention factor of the wavelet coefficients . Based on these issues , created a program to analyze the influence of variations in the level of wavelet decomposition and coefficient of variation coefficient of cutting a cut on the image . The object of this final project is 30 iris image that has been converted into polar form presented . Having experienced the process of cutting coefficients , to prevent the image of the reverse transformation has a big difference to the original image , in this study was calculated Mean Square Error ( MSE), Peak Signal to Noise Ratio ( PSNR), and Euclidean distance to determine the level of similarity of image input and output images . The Highest values of Retention ​, MSE and the Euclidean distance is obtained at the level of decomposition of 1 and the lowest at the level of decomposition 6 . While the truncated coefficient and PSNR values ​​obtained at the highest level of decomposition of 6 and the lowest level of decomposition of 6 . The variation coefficient of pieces , value retention and highest PSNR obtained at koefsien piece 5 and the lowest coefficient of 50 pieces . While the coefficient value is truncated , MSE and the highest Euclidean distance is obtained at koefsien pieces 50 and the lowest coefficient of 5 pieces . Keywords: iris, texture analysis, Haar wavelet transform, MSE, PSNR, E uclidean distance

Genetic Algorithm (GA) has been successfully applied to codebook design for vector quantization and its candidate solutions are normally turned by LBG algorithm. In this paper, to solve premature phenomenon and falling into local optimum of GA, a new Genetic Simulated Annealing-based Kernel Vector Quantization (GSAKVQ) is proposed from a different point of view. The simulated annealing (SA) method proposed in this paper can approach the optimal solution faster than the other candidate approaches. In the frame of GA, firstly, a new special crossover operator and a mutation operator are designed for the partition-based code scheme, and then a SA operation is introduced to enlarge the exploration of the proposed algorithm, finally, the Kernel function-based fitness is introduced into GA in order to cluster those datasets with complex distribution. The proposed method has been extensively compared with other algorithms on 17 datasets clustering and four image compression problems. The experimental results show that the algorithm can achieve its superiority in terms of clustering correct rate and peak signal-to-noise ratio (PSNR), and the robustness of algorithm is also very good. In addition, we took “Lena” as an example and added Gaussian noise into the original image then adopted the proposed algorithm to compress the image with noise. Compared to the original image with noise, the reconstructed image is more distinct, and with the parameter value increasing, the value of PSNR decreases.

Motion estimation (ME) is a process of correlating the frames by pointing the blocks in the candidate frame similar to those in the previous frame. Hexagonal Search (HS) reduces 10–15% search points over Diamond Search (DS) with negligible Peak Signal to Noise Ratio (PSNR) loss. Adapting an optimization will achieve a massive improvement on the search process. The proposed HS-GA method combines the HS with genetic algorithm (GA) to optimize ME. The HS-GA method reduces 72% on average search points pre block (ASPB) with PSNR loss of 2.5 dB and single generation. If the GA is executed with 10 generations the PSNR loss is reduced to 0.48 dB, but ASPB drastically increase to ∼316%. The proposed work aims at multi objective on selection of better chromosomes and on control of generations to avoid the unnecessary generations to improve the GA. This process is named HSIGA(HS with Intelligent GA). The proposed HSIGA adapts five intelligent techniques like Histogram to decide new generations, Region based chromosome selection, Mutation on inner search gene, second rank valley point stopping criteria and interlaced HS to improve search process with negligible PSNR loss with in affordable search.

In this paper, an enhanced genetic algorithm (EGA) technique-based fractional order PID controller (FOPID) for simultaneous control of synchronous buck converters is proposed. The novelty of EGA method is that enhancement process is done by the Rat swarm algorithm (RSA). The computational scheme of RSA is used to enhance the performance of traditional genetic algorithm (GA) and a modified version of GA, which is used for optimal tuning of FOPID controller. In the proposed methodology, the FOPID controller gain is adjusted to achieve the preferred response that is demonstrated by EGA technique. The objective of the proposed technique is “the smooth DC output voltage is attained under damped response”. The proposed control method is executed and simulated in MATLAB platform and the performance is analyzed with the existing methods. Statistical analysis of mean, median and standard deviation for the proposed and existing methods is also analyzed. The proposed FOPID controller produces regulated output DC voltage having 34% lesser overshoots and two times faster settling time when compared to the existing dynamic evolution controller for Synchronous Buck Converters (SBCs). The designed optimal FOPID controlled SBC is also implemented in hardware using Atmega16A processor for different loading conditions for better applications.

Multiclass SVM based adaptive filter for removal of high density impulse noise from color images

In this paper, a comparative study of different wavelet filters using improved sub-band adaptive thresholding function for denoising of satellite images, based on evolutionary algorithms, has been performed. In this approach, the stochastic global optimisation techniques such as Cuckoo Search (CS) algorithm, artificial bee colony (ABC) and particle swarm optimisation (PSO) are used for obtaining the parameters of adaptive thresholding function required for optimum performance. The visual and quantitative results clearly show the increased efficiency and flexibility of the proposed CS algorithm based on Meyer wavelet filter over various other wavelet filters for image denoising. From the comparative study of different wavelet filters, it is found that the proposed Meyer wavelet-based CS algorithm denoising approach gives better performance in terms of signal-to-noise ratio (SNR), peak signal-to-noise ratio (PSNR), mean square error (MSE) and mean as compared to ABC- and PSO-based denoising approach. The proposed technique has been tested on several satellite images. The quantitative (EKI or EPI, mean, MSE, SNR and PSNR) and visual (denoised images) results show the superiority of the proposed technique over conventional and state-of-art image denoising techniques.

Rate control plays a key role in a video encoder, and it has been a challenging research topic. In this paper, an adaptive variable length group of picture (GOP) level rate control algorithm for H.264 based on scene change detection is proposed. In the proposed algorithm, the length of current GOP changes adaptively with the occurrence of scene change, so that the entire sequence contains a fixed number of I frame to avoid the actual bit rate increasing. Furthermore, it adjusts the QP of the scene change frame to ease the pressure of the buffer. Simulation results show that the proposed algorithm improves the video quality significantly on the basis of meeting bit rate constraints accurately, which shows better adaptability for sequences with scene change occurrence frequently. When compared with existed schemes, the average peak signal-to-noise ratio (PSNR) improvement is more than 1.4 dB.

In this paper, we propose several adaptive methods to improve the fractal block coding (FBC) of still images. First of all, the range of the contrast scaling is adaptively selected according to its statistical property: then, we search for the best quantized value in the selected range. Then, the domain pool of the current range block is predicted using a finite-state method, which uses the information of previous best-match domain blocks and domain pools. The size of the domain pool is automatically adjusted; however, conventional studies can only determine a fixed-size domain pool. Therefore, the search process for finding the best-match domain block in our approach is more flexible. In addition, we propose a finite-state method to estimate the mean (average pixel value) of the current range block by using the means of the previous blocks. From the simulation, we achieve 0.35 dB improvement in peak signal-to-noise ratio and 10.5% reduction in bit rate on average compared with the basic FBC scheme. Finally, the encoding and decoding time of different FBC schemes is also evaluated for completeness.

Histogram equalization is a well known technique used for contrast enhancement. However, it changes the brightness of an image and hence, not suitable for consumer electronic products. This paper proposes a new technique, Bi Level Histogram Modification using Genetic Algorithm (BLHM) for the purpose of enhancing the contrast as well as preserving details for any given input image. The basic idea of this technique is partition the input image histogram into two sub-histograms based on its mean. Then, a bi-criteria optimization problem is formulated to satisfy the aforementioned requirements and the two sub-histograms are modified by selecting the optimal contrast enhancement parameters. Finally, the union of two modified sub-histograms produces a contrast enhanced and details preserved output image. While formulating the optimization problem, Genetic Algorithm is employed to control the degree of enhancement. This mechanism enhances the contrast of the input image better than its existing contemporary HE methods. The performance of the proposed method is measured in terms of Discrete Entropy and Contrast Improvement Index.

Objective To assess image quality and optimal keV levels of spectral CT virtual monochromatic series in atypical vertebral osteoblastic metastases of lung cancer. Methods Patients with pathologically proven lung cancer who had undergone spectral CT (Discovery CT750 HD, GE) of the chest or abdomen more than twice were analyzed retrospectively. Vertebral osteoblastic metastases were confirmed by follow-up CT examinations, biopsy ,emission CT, PET/CT. The bone lesions were missed due to atypicalappearance or presented with suspicious density at the first exmaination were chosen. A total of twenty two patients with 72 lesions were enrolled. Conventional polychromatic images (140 kVp) and monochromatic images were generated (40 to 140 keV at an interval of 10 keV). The signal to noise ratio(SNR) and contrast to noise ratio (CNR) of the lesions were measured and calculated at all the 12 series images by two radiologists. The two radiologists scored the different series of images independently. Paired sample t test was used to compare SNR and CNR among 40 to 140 keV and 140 kVp images. The inter-observer agreement between the two radiologists was evaluated using Kappa statistics and the difference of different images intrarobserver was compared using Wilcoxon signed rank test. Results The SNR of 140 kVp was 48.03±20.17, and those of 50, 60 70 keV images were 58.52±22.92, 64.07±24.88, 59.54±22.04. SNR of 50, 60, 70 keV images was higher than that of 140 kVp (t values were 13.366,16.278, 18.889,P<0.01), 60 keV images had the highest SNR. The CNR of 140 kVp was 7.64±4.52 and those of 50, 60, 70 and 80 keV images were 7.95±4.99, 8.12±5.08, 8.16±5.05, 7.89±4.81. CNR of 50, 60, 70 and 80 keV images were higher than that of 140 kVp image (t values were 3.519, 5.050, 6.231, 4.413, P<0.01), 60 keV and 70 keV images had the higher CNR than 50 keV and 80 keV. Good interobserver agreement was found for the evaluation of image quality (Kappa =0.864,P<0.05).Moreover, the scores of both the two observes for 60 keV images were higher than 140 kVp images and higher than other monochromatic images. Conclusion Atypical vertebral osteoblastic metastases of lung cancer are more conspicuous in virtual monochromatic images as compared wtih 140 kVp polychromatic images and the optimal energy level was 60 keV, repectively. Key words: Lung neoplasms; Bone neoplasms; Neoplasm metastasis; Tomography, X-ray computed

Medical images are often of low contrast and noisy (lack of clarity) due to the circumstances they are being captured. De-noising of ultrasound images is a difficult task as compare to other medical images because of noise, blurring of edges and artifacts. The Bayesian shrinkage method has been selected for thresholding based on its sub-band dependency property. The spatial domain based de-noising filtering techniques, using soft thresholding method are compared with the proposed method using Genetic Algorithm (GA). A proposed technique includes GA and results are compared with existing spatial domain based de-noising filtering techniques. The proposed algorithm provides enhanced visual clarity for diagnosing the medical images. The proposed method based on GA assesses the better performance on the basis of the quantitative metric like Peak Signal-to-Noise Ratio (PSNR) and Fitness value. The overall simulated result shows that proposed technique outperforms the prevailing de-noising filtering methods in terms of edge preservation and visuality.

In the world of image processing domain, removal of high density impulse noise is always a prominent area for research. In this paper, an efficient Modified Decision based unsymmetric trimmed median filter algorithms for the removal of impulse noise has been proposed with color images rather than gray scale images by separation red- green- blue plane of color image. The performance of the system is analyzed in terms of Mean square error (MSE), Peak signal to noise ratio (PSNR) image enhancement factor (IEF) and time required for executing the algorithms for different noise densities. Simulation results shows that proposed algorithm outperforms the existing algorithms even at high noise densities for color images. Many experiments are conducted to validate efficiency of the proposed algorithm.

Assembly line balancing (ALB) is used in many industries to minimize the number of stations, improve the efficiency and work load balance among stations. Enhanced Genetic Algorithm (EGA) is proposed using precedence preservative crossover and scramble mutation techniques, aiming to minimize two objectives; number of stations as a primary objective and smoothing index as a secondary objective. Benchmark problems were selected from the literature, used to test the efficacy of the algorithm and to compare the results with the well-known algorithm SALOME. The results showed high efficacy while handling two objectives. It outperformed SALOME in work load balance and efficiency. On the other, EGA outperformed the Genetic Algorithm (GA) developed by [1] in minimizing the number of stations.

BackgroundMining epistatic loci which affects specific phenotypic traits is an important research issue in the field of biology. Bayesian network (BN) is a graphical model which can express the relationship between genetic loci and phenotype. Until now, it has been widely used into epistasis mining in many research work. However, this method has two disadvantages: low learning efficiency and easy to fall into local optimum. Genetic algorithm has the excellence of rapid global search and avoiding falling into local optimum. It is scalable and easy to integrate with other algorithms. This work proposes an epistasis mining approach based on genetic tabu algorithm and Bayesian network (Epi-GTBN). It uses genetic algorithm into the heuristic search strategy of Bayesian network. The individual structure can be evolved through the genetic operations of selection, crossover and mutation. It can help to find the optimal network structure, and then further to mine the epistasis loci effectively. In order to enhance the diversity of the population and obtain a more effective global optimal solution, we use the tabu search strategy into the operations of crossover and mutation in genetic algorithm. It can help to accelerate the convergence of the algorithm.ResultsWe compared Epi-GTBN with other recent algorithms using both simulated and real datasets. The experimental results demonstrate that our method has much better epistasis detection accuracy in the case of not affecting the efficiency for different datasets.ConclusionsThe presented methodology (Epi-GTBN) is an effective method for epistasis detection, and it can be seen as an interesting addition to the arsenal used in complex traits analyses.

Objective To compare the rat head MRI image quality of small aperture orthogonal phased-array creatural experimental coil and the C3 surface coil.Methods Ten male SD rats underwent both coils for head MRI.The signal intensities and background noise values of the cerebral cortex,cerebellum, brain stem, muscle and aqueous humor in eye were measured, and signal to noise ratio (SNR)was calculated.Image quality scores were obtained from 2 radiologists, t test was performed for statistical differences.Results SNR values over the cerebral cortex, cerebellum, brain stem, muscle and aqueous humor in eye were obtained from the images of all protocols.Statistically significant differences of the two coils were achieved in the cerebral cortex in the transverse T1WI (SNR sense-body=41.3±23.5; SNR C3=10.3±0.5; t=-3.21), sagittal T2WI (SNR sense-body=63.8±16.6; SNR C3=37.9±4.4; (=-4.00) and coronal T2WI (SNR sense-body=91.6±23.8, SNR C3=38.5±11.8; t=-5.69)(P＜0.01), so were in cerebellum, brain stem and aqueous humor in eye.Statistical differences of muscular tissue were observed merely in the transverse axial TiWI (SNR sense-body=39.9±21.1; SNR C3=8.3±1.7; t=-3.64) and the coronal T2WI (SNR sense-body=23.5±9.8; SNR C3=1 1.9±4.1; t=-3.10)(P＜0.01).Coronal T2WI score achieved the top in the small aperture orthogonal phased array coil, and the image qualities of diffusion-weighted imaging(DWI)in both two coils were low.Conclusion In head MRI of rats, the SNR value and head image quality of small aperture orthogonal phased-array coil is superior to that of the frequently used C3 surface coil. Key words: Magnetic resonance imaging; Small aperture orthogonal phased-array coil; Surface coils; Rat; Image quality