One-Class Classification Using lp-Norm Multiple Kernel Fisher Null-Space

The paper addresses the one-class classification (OCC) problem and advocates a one-class multiple kernel learning (MKL) approach for this purpose. To this aim, based on the Fisher null-space one-class classification principle, we present a multiple kernel learning algorithm where an $\ell_p$-norm constraint ($p\geq1$) on kernel weights is considered. We cast the proposed one-class MKL task as a min-max saddle point Lagrangian optimisation problem and propose an efficient method to solve it. An extension of the proposed one-class MKL approach is also considered where several related one-class MKL tasks are learned concurrently by constraining them to share common kernel weights. <br>An extensive assessment of the proposed method on a range of data sets from different application domains confirms its merits against the baseline and several other algorithms.<br>

One-Class Classification Using lp-Norm Multiple Kernel Fisher Null-Space

The paper addresses the one-class classification (OCC) problem and advocates a one-class multiple kernel learning (MKL) approach for this purpose. To this aim, based on the Fisher null-space one-class classification principle, we present a multiple kernel learning algorithm where an $\ell_p$-norm constraint ($p\geq1$) on kernel weights is considered. We cast the proposed one-class MKL task as a min-max saddle point Lagrangian optimisation problem and propose an efficient method to solve it. An extension of the proposed one-class MKL approach is also considered where several related one-class MKL tasks are learned concurrently by constraining them to share common kernel weights. <br>An extensive assessment of the proposed method on a range of data sets from different application domains confirms its merits against the baseline and several other algorithms.<br>

Removal of Gaussian White Noises from the image by probability map prediction based Deep learning approach

Deep learning methods have gained an increasing research interest, especially in the field of image denoising. Although there are significant differences between the different types of deep learning techniques used for natural image denoising, it includes significant process and procedure differences between them. To be specific, discriminative learning based on deep learning convolutional neural network (CNN) may effectively solve the problem of Gaussian noise. Deep learning based optimization models are useful in predicting the true noise level. However, no relevant research has attempted to summarize the different deep learning approaches for performing image denoising in one location. It has been suggested to build the proposed framework in parallel with the previously trained CNN to enhance the training speed and accuracy in denoising the Gaussian White Noise (GWN). In the proposed architecture, ground truth maps are created by combining the additional patches of input with original pictures to create ground truth maps. Furthermore, by changing kernel weights for forecasting probability maps, the loss function may be reduced to its smallest value. Besides, it is efficient in terms of processing time with less sparsity while enlarging the objects present in the images. As well as in conventional methods, various performance measures such as PSNR, MSE, and SSIM are computed and compared with one another.

Learning to Diversify for Ensembling of An Amount of Classifiers

Not only common man but also intelligent machine always merge all available decisions to solve problems. However, given an amount of learned classifiers, how to select and combine diverse classifiers for machine is still a grand challenge in the literature for decades of history. In this paper, we introduce a novel approach for classifier ensemble, learning to diversify, which learns to adaptively combine classifiers by considering both accuracy and diversity. Specifically, our approach, learning to diversify via weighted kernels, performs classifier combination by optimizing a direct but simple criterion: maximizing ensemble accuracy and adaptive diversity simultaneously by minimizing a loss function. Given a measure formulation, the diversity is calculated with weighted kernels, i.e., the diversity is measured on the component classifiers’ outputs which are kernelized and weighted. Moreover, we propose an iterative training algorithm for weights optimization, where this loss function is iteratively minimized by estimating the kernel weights in conjunction with the classifier weights. Extensive experiments on a variety of classification benchmark datasets show that the proposed approach consistently outperforms state-of-the-art ensembles.

DETERMINATION OF SUITABLE WINTER CHICKPEA (CICER ARIETINUM L.) CULTIVARS UNDER KAHRAMANMARAŞ CONDITIONS

This study was conducted with the chickpea cultivars of Işık-05, Azkan, Sarı 98, Hisar, Çakır, Aydın 92, Yaşa-05, Menemen 92, Cevdetbey, Çağatay, Aksu and two local cultivars over the experimental fields of Kahramanmaraş Eastern Mediterranean Transitional Zone Agricultural Research of Institute in 2014-2015 cropping years. Experiments were conducted in randomized blocks design with 3 replications. Quality traits of plant height, the first pod height, number of branches per plant, number of pods per plant, number of kernels per plant, kernel weight per plant, kernel yield, 100-kernel weights were investigated. The differences in plant height, the first pod height, number of branches per plant, number of pods per plant, number of kernels per plant, kernel weight per plant, kernel yield and 100-kernel weight of the genotypes were found to be significant. Kernel yields of the genotypes varied between 425.40 - 267.93 kg da-1 with the greatest value from Çakır cultivar and the lowest value from Hisar cultivar.

Estimation of conditional distribution functions from data with additional errors applied to shape optimization

We study the problem of estimating conditional distribution functions from data containing additional errors. The only assumption on these errors is that a weighted sum of the absolute errors tends to zero with probability one for sample size tending to infinity. We prove sufficient conditions on the weights (e.g. fulfilled by kernel weights) of a local averaging estimate of the codf, based on data with errors, which ensure strong pointwise consistency. We show that two of the three sufficient conditions on the weights and a weaker version of the third one are also necessary for the spc. We also give sufficient conditions on the weights, which ensure a certain rate of convergence. As an application we estimate the codf of the number of cycles until failure based on data from experimental fatigue tests and use it as objective function in a shape optimization of a component.

An Autocontouring Method for the Kidneys Using an Adaptive Weighted Multikernel Support Vector Machines

In radiotherapy, the location of the target area is very important. If the target area is small, the treatment is not complete, so the location of the target area is generally larger than the actual cancerous site. However, the damage of radiotherapy to normal cells is the same. In order to reduce the damage to the body as much as possible, we need to complete the most suitable target area. This paper uses an adaptive weighted multikernel support vector machine, which solves the parameter problem in the traditional multikernel support vector machine. The new AW-SVM can adjust the kernel weights adaptively. We completed our experiment on the abdominal MR dataset, using DSI as an evaluation indicator, and the experimental results showed its excellent classification performance. The minimum value of DSI in all results is 0.9654 (more than 0.7 is acceptable).

Agronomic and Morpho-Physiological Characterization of Some Advanced Lines being Selected in Durum Wheat Cultivated in Semi-Arid Conditions

Knowledge of agro-physiological traits associated with drought tolerance would be useful for developing breeding materials for drought-prone environments. This study was conducted to estimate genetic variability among nine durum wheat genotypes in response to drought. Our results indicated that the effect of the campaign, genotype, and genotype × interaction was significant for the thirteen variables measured, except for the relative water content. The variability observed was greater for grain yield, biomass, ear fertility, straw and economic yields, chlorophyll content, and cell integrity. Heritability was high for the number of grains per ear and the chlorophyll content; medium for thousand kernel weights, low for grain yield, biomass, and economic yield, and zero for the rest of the variables measured. The results also showed that the agro-morphological characters were significantly linked to each other, unlike the physiological characters which showed non-significant relation between them and with the agro-morphological characters. This suggests that among the varieties evaluated, the selection of those which are tolerant and with high yield potential should therefore be made on a case-by-case basis and not based on a specific physiological character, a marker of tolerance, highly correlated with yield grain. The nine varieties evaluated were subdivided into three divergent clusters of three varieties each. Cluster C1 consists of the least performing varieties, unlike the other two clusters which bring appreciable gains for several characteristics including grain yield, biomass, the weight of 1000 grains, straw yield, and ear fertility and a marked improvement in chlorophyll content and a significant reduction in damage to the cell membrane by thermal stress. In conclusion and following their divergence, it is suggested to use the varieties of clusters C2 and C3 in crossing with the varieties of cluster C1 to improve and reconcile stress tolerance and yield potential in the same genetic background.

Multiple Kernel Graph Cut for SAR Image Change Detection

Complementary information between two difference images (DI’s) has great contribution to improve change detection performances. Based on the effectiveness and flexibility of the multiple kernel learning (MKL) in information fusion, we develop a multiple kernel graph cut (MKGC) algorithm for synthetic aperture radar (SAR) image change detection. An energy function containing a weighted summation kernel is proposed for fusing the complementary information between the subtraction image and the ratio image. By iteratively minimizing the energy function, the kernel weights, region parameters and region labels are estimated automatically and optimally. Besides of it avoids modeling, MKGC also has a complete description of the changed areas and the strong noise immunity. Experiments on real GaoFen-3 SAR data set demonstrate the effectiveness of the MKGC algorithm, and illustrate that it is a good candidate for SAR image change detection.