Non-dominated sorting genetic algorithm — II supported neural network in classifying forest types
01 Apr 2017-pp 1-6
TL;DR: Experimental results have indicated that the proposed NN-NSGA-II model is superior to the GA-NN model to a greater extent.
Abstract: Pixel classification in land scape images has been found to be challenging. The problem becomes more challenging in forest images due to the similar spectral features of pixels situated close to each other. Geographically weighted variables have been employed to classify the two different species namely Cryptomeria japonica (Japanese Cedar or Sugi) and Chamaecyparisobtusa (Japanese Cypress or Hinoki) and one mixed forest class. Previous attempts have shown reasonable improvement in this task using Genetic Algorithm supported Neural Network over other traditional approaches. Motivated by this, a NSGA — II supported Neural Network (NN-NSGA — II) classifier is proposed. The proposed model has been compared with GA-NN (ANN trained with Genetic Algorithm with a single objective function) classifiers in terms of confusion matrix based performance metrics such as accuracy, precision, recall and F-Measure. Experimental results have indicated that the proposed NN-NSGA-II model is superior to the GA-NN model to a greater extent.
Citations
More filters
01 Aug 2017
TL;DR: Experimental results clearly show the superiority of the proposed NN-NSGA-II model with different features, which has been evaluated using various performances measuring metrics such as accuracy, precision, recall and F-measure.
Abstract: Automated, efficient and accurate classification of skin diseases using digital images of skin is very important for bio-medical image analysis. Various techniques have already been developed by many researchers. In this work, a technique based on meta-heuristic supported artificial neural network has been proposed to classify images. Here 3 common skin diseases have been considered namely angioma, basal cell carcinoma and lentigo simplex. Images have been obtained from International Skin Imaging Collaboration (ISIC) dataset. A popular multi objective optimization method called Non-dominated Sorting Genetic Algorithm — II is employed to train the ANN (NNNSGA-II). Different feature have been extracted to train the classifier. A comparison has been made with the proposed model and two other popular meta-heuristic based classifier namely NN-PSO (ANN trained with Particle Swarm Optimization) and NN-GA (ANN trained with Genetic algorithm). The results have been evaluated using various performances measuring metrics such as accuracy, precision, recall and F-measure. Experimental results clearly show the superiority of the proposed NN-NSGA-II model with different features.
39 citations
01 Oct 2017
TL;DR: Experimental results indicated towards the superiority of the proposed bag-of-features enabled NN-NSGA-II model in terms of testing phase confusion matrix based performance measuring metrics.
Abstract: The current work proposes a neural based detection method of two different skin diseases using skin imaging. Skin images of two diseases namely Basel Cell Carcinoma and Skin Angioma are utilized. SIFT feature extractor has been employed followed by a clustering phase on feature space in order to reduce the number of features suitable for neural based models. The extracted bag-of-features modified dataset is used to train metaheuristic supported hybrid Artificial Neural Networks to classify the skin images in order to detect the diseases under study. A well-known multi objective optimization technique called Non-dominated Sorting Genetic Algorithm — II is used to train the ANN (NN-NSGA-II). The proposed model is further compared with two other well-known metaheuristic based classifier namely NN-PSO (ANN trained with PSO) and NN-CS (ANN trained with Cuckoo Search) in terms of testing phase confusion matrix based performance measuring metrics such as accuracy, precision, recall and F-measure. Experimental results indicated towards the superiority of the proposed bag-of-features enabled NN-NSGA-II model.
36 citations
TL;DR: A modified Flower Pollination Algorithm has been employed to train Artificial Neural Network to predict soil moisture quantity and the proposed method is compared with well known PSO supported ANN and Cuckoo Search supported ANN along with MLP-FFN classifier.
Abstract: Predicting soil moisture quantity could directly help the people engaged in sustainable agriculture and associated socio-economic structures Recently researchers have engaged traditional and machine learning based models to predict soil moisture quantity In the current study a modified Flower Pollination Algorithm (MFPA) has been employed to train Artificial Neural Network (ANN) to predict soil moisture quantity The proposed method is compared with well known PSO (Particle Swarm optimization) supported ANN and Cuckoo Search (CS) supported ANN along with MLP-FFN classifier The stability of the proposed model in presence of varying weather conditions has been established by performing a stability analysis using data level perturbation Experimental results have indicated that NN-MFPA achieved an average RMSE of 00019 and outperformed other models The ingenuity of the proposed model is further established by performing Wilcoxon rank test with 5% level of significance
32 citations
01 Aug 2017
TL;DR: This work describes an method for biomedical image enhancement using modified Cuckoo Search Algorithm with some Morphological Operation and a new technique has been proposed to enhance biomedical images using modified cuckoo search algorithm and morphological operation.
Abstract: This work describes an method for biomedical image enhancement using modified Cuckoo Search Algorithm with some Morphological Operation. In recent years, various digital image processing techniques are developed. Computer Vision, machine interfaces, manufacturing industry, data compression for storage, vehicle tracking and many more are some of the domains of digital image processing application. In most of the cases, digital biomedical images contains various types of noise, artifacts etc. and are not useful for direct applications. Before using it in any process, the input image has to be gone through some preprocessing stages; such preprocessing is generally called as image enhancement. In this work, a new technique has been proposed to enhance biomedical images using modified cuckoo search algorithm and morphological operation. Presence of noise and other unwanted objects generates distortion in an image and it will affect the ultimate result of the process. In case of biomedical images, accuracy of the results is very important. It may also decrease the discernibility of many features inside the images. It can affect the classification accuracy. In this work, this issue has been targeted and improved by obtaining better contrast value after converting the color image into grayscale image. The basic property of the cuckoo search algorithm is that the amplitudes of its components are capable to objectively describe the contribution of the gray levels to the formation of image information for the best contrast value of a digital image. The proposed method modified the conventional cuckoo search method by employing the McCulloch's method for levy flight generation. After computing the best contrast value, morphological operation has been applied. In morphological operation based phase, the intensity parameters are tuned for quality enhancement. Experimental results illustrate the effectiveness of this work.
32 citations
01 Oct 2017
TL;DR: A gradient-based blood vessel segmentation technique is proposed to assist retinal image analysis and to extract the retinal vessels and it outperformed the corresponding values obtained by the other standard edge detectors, namely Sobel, Prewitt, Canny, and Robert's.
Abstract: Image segmentation is one of the major research domains in several applications including retinal blood vessel segmentation, which is an active research area. Vasculature structure analysis is an interesting and effective method for disease detection and analysis. In this work, a gradient-based blood vessel segmentation technique is proposed to assist retinal image analysis and to extract the retinal vessels. Edge detection is considered one of the major steps in the present work to characterize the boundaries. Itis used to reduce the unusual information and to preserve the necessary structural information. Various filters are constructed to gradient computation and edge detection. In the current work, a new method along with a new filter (kernel) has been proposed to detect edges efficiently. The results are compared with some well-known kernels. The proposed approach achieved Pratt Score 99.1536 value, which outperformed the corresponding values obtained by the other standard edge detectors, namely Sobel, Prewitt, Canny, and Robert's.
26 citations
References
More filters
Posted Content•
TL;DR: E elegant connections between the concepts of Informedness, Markedness, Correlation and Significance as well as their intuitive relationships with Recall and Precision are demonstrated.
Abstract: Commonly used evaluation measures including Recall, Precision, F-Measure and Rand Accuracy are biased and should not be used without clear understanding of the biases, and corresponding identification of chance or base case levels of the statistic. Using these measures a system that performs worse in the objective sense of Informedness, can appear to perform better under any of these commonly used measures. We discuss several concepts and measures that reflect the probability that prediction is informed versus chance. Informedness and introduce Markedness as a dual measure for the probability that prediction is marked versus chance. Finally we demonstrate elegant connections between the concepts of Informedness, Markedness, Correlation and Significance as well as their intuitive relationships with Recall and Precision, and outline the extension from the dichotomous case to the general multi-class case.
5,092 citations
01 Oct 2002
TL;DR: The assumption that the class imbalance problem does not only affect decision tree systems but also affects other classification systems such as Neural Networks and Support Vector Machines is investigated.
Abstract: In machine learning problems, differences in prior class probabilities -- or class imbalances -- have been reported to hinder the performance of some standard classifiers, such as decision trees. This paper presents a systematic study aimed at answering three different questions. First, we attempt to understand the nature of the class imbalance problem by establishing a relationship between concept complexity, size of the training set and class imbalance level. Second, we discuss several basic re-sampling or cost-modifying methods previously proposed to deal with the class imbalance problem and compare their effectiveness. The results obtained by such methods on artificial domains are linked to results in real-world domains. Finally, we investigate the assumption that the class imbalance problem does not only affect decision tree systems but also affects other classification systems such as Neural Networks and Support Vector Machines.
2,830 citations
"Non-dominated sorting genetic algor..." refers background in this paper
...Several studies revealed that performance comparison in terms of accuracy is not sufficient to claim the ingenuity of a model [34, 36-37]....
[...]
Book•
12 Jul 1996
TL;DR: The authors may not be able to make you love reading, but neural networks a systematic introduction will lead you to love reading starting from now.
Abstract: We may not be able to make you love reading, but neural networks a systematic introduction will lead you to love reading starting from now. Book is the window to open the new world. The world that you want is in the better stage and level. World will always guide you to even the prestige stage of the life. You know, this is some of how reading will give you the kindness. In this case, more books you read more knowledge you know, but it can mean also the bore is full.
2,278 citations
"Non-dominated sorting genetic algor..." refers background in this paper
...The perceptron learning rule is assures optimal weight vector for ANN in finite number of iterations [32]....
[...]
TL;DR: A critical review of the most important evolutionary-based multiobjective optimization techniques developed over the years, emphasizing the importance of analyzing their Operations Research roots as a way to motivate the development of new approaches that exploit the search capabilities of evolutionary algorithms.
Abstract: This paper presents a critical review of the most important evolutionary-based multiobjective optimization techniques developed over the years, emphasizing the importance of analyzing their Operations Research roots as a way to motivate the development of new approaches that exploit the search capabilities of evolutionary algorithms Each technique is briefly described with its advantages and disadvantages, its degree of applicability and some of its known applications Finally, the future trends in this discipline and some of the open areas of research are also addressed
1,328 citations
"Non-dominated sorting genetic algor..." refers background in this paper
...Multi-objective optimization can formally be framed as follows [28]: Find the vector ⃗ = [ , , ....
[...]
TL;DR: The results indicate that the performance of the univariate DT is acceptably good in comparison with that of other classifiers, except with high-dimensional data, and the use of attribute selection methods does not appear to be justified in terms of accuracy increases.
Abstract: Choice of a classification algorithm is generally based upon a number of factors, among which are availability of software, ease of use, and performance, measured here by overall classification accuracy. The maximum likelihood (ML) procedure is, for many users, the algorithm of choice because of its ready availability and the fact that it does not require an extended training process. Artificial neural networks (ANNs) are now widely used by researchers, but their operational applications are hindered by the need for the user to specify the configuration of the network architecture and to provide values for a number of parameters, both of which affect performance. The ANN also requires an extended training phase. In the past few years, the use of decision trees (DTs) to classify remotely sensed data has increased. Proponents of the method claim that it has a number of advantages over the ML and ANN algorithms. The DT is computationally fast, make no statistical assumptions, and can handle data that are represented on different measurement scales. Software to implement DTs is readily available over the Internet. Pruning of DTs can make them smaller and more easily interpretable, while the use of boosting techniques can improve performance. In this study, separate test and training data sets from two different geographical areas and two different sensors—multispectral Landsat ETM+ and hyperspectral DAIS—are used to evaluate the performance of univariate and multivariate DTs for land cover classification. Factors considered are: the effects of variations in training data set size and of the dimensionality of the feature space, together with the impact of boosting, attribute selection measures, and pruning. The level of classification accuracy achieved by the DT is compared to results from back-propagating ANN and the ML classifiers. Our results indicate that the performance of the univariate DT is acceptably good in comparison with that of other classifiers, except with high-dimensional data. Classification accuracy increases linearly with training data set size to a limit of 300 pixels per class in this case. Multivariate DTs do not appear to perform better than univariate DTs. While boosting produces an increase in classification accuracy of between 3% and 6%, the use of attribute selection methods does not appear to be justified in terms of accuracy increases. However, neither the univariate DT nor the multivariate DT performed as well as the ANN or ML classifiers with high-dimensional data.
1,013 citations
"Non-dominated sorting genetic algor..." refers background in this paper
...Thus, several works [1-8] have attempted the problem and reported reasonable accuracy....
[...]