Leaf recognition systems can be used for automatic plant taxonomy and provide understanding managing of plants in botany, medicine, industry food sector. Trees flowery classified by using leaf recognition. This paper proposes a simple method based on bisection leaves After preprocessing techniques are applied leaves, 7 low-cost morphological features extracted which the literature. We produced 3 additional half images. Most species have structure that resembles each other lot. For these...

We are introducing a new design goal called area priority to determine optimal sensor node distribution. The environment in which the wireless network (WSN) will be placed is divided into parts and priorities attached these parts. Priorities make deployment problem adaptable nonhomogeneous environments with regions that have different importance levels such as forests. Various tree/animal types densities, residential forest can classified by concept we propose. also develop genetic...

The Traveling Salesman Problem (TSP) is one of the extensively studied combinatorial optimization problems. Various exact or approximation algorithms are devised for solving Euclidean TSP that determine shortest route through a given set points in 3-dimensional space. In this paper, we proposed genetic algorithm-based solution where all on surface sphere. A Java-based interactive visualization tool also developed using Java 3D and results obtained different problem sizes presented.

Extraction of flower regions from complex background is a difficult task and it an important part image retrieval recognition. In this article, we propose Ant Colony Optimization (ACO) algorithm as general color clustering method, test on images case study object boundary extraction. The segmentation methodology consists six steps: space conversion, generation candidate cluster centers, ant colony optimization method to select optimum merging centers which are close each other, by...

Abstract Although neural networks (NN) are important especially for engineers, scientists, mathematicians and statisticians, they may also be hard to understand. In this article, application areas of NN discussed, basic components described it is explained how an work. A web‐based simulation visualization tool (EasyLearnNN) developed using Java 2D teaching concepts. Perceptron, ADALINE, Multilayer LVQ SOM models related training algorithms implemented. As a result, comparison with other...

Outlier detection is considered as one of the crucial research areas for data mining. Many methods have been studied widely and utilized achieving better results in outlier from existing literature; however, effects these few ways are inadequate. In this paper, a stacking-based ensemble classifier has proposed along with four base learners (namely, Rotation Forest, Random Bagging Boosting) Meta-learner Logistic Regression) to progress performance. The mechanism evaluated on five datasets...

Deployment in wireless sensor networks (WSN) addresses maximising the coverage of sensors and reducing total cost deployment. The area‐priority concept for WSN deployment that authors contributed to literature recently allows environments with regions have different importance or priority levels. In this study, propose first priority‐estimation method area‐priority‐based deployments. First, a satellite image environment will be used is clustered by K ‐means algorithm using colour features...

In this study, ensemble learning based image classification method is proposed by using both features extracted means of pre-trained convolutional neural networks (CNN) and hand-crafted. Recently, deep models have been widely used in computer vision applications significantly increase performance. scope, process performed adding 4 hand-crafted to 4096 on the CIFAR-10 dataset. The contribution performance system measured together. Classification accuracy rate as criterion. Experimental...

In this paper, a new license plate information retrieval system is designed and developed. The has two main modules: segmentation recognition. segmentation, interested on the image extracted through processes of Kaiser resizing, morphological filtering, artificial shifting bi-directional vertical thresholding. recognition module, novel approach for principal component analysis (PCA) fast backpropagation neural net composition used as recognizer. about construction Eigen space PCA that...

Outliers are abnormal, aberrant or outlying observations in data and can cause distortion of estimations statistical models. Identification outliers is an important process for preventing faulty conclusions analysis. Simultaneous outlier detection, which genetic algorithms (GA) provide, more successful than the methods based on detecting one by when order detection important. In this study, we derived new approaches information criteria Akaike's criterion (AIC) Bozdogan's complexity (ICOMP)...

Glioma is one of the most common tumors brain. The detection and grading glioma at an early stage very critical for increasing survival rate patients. Computer-aided (CADe) computer-aided diagnosis (CADx) systems are essential important tools that provide more accurate systematic results to speed up decision-making process clinicians. In this paper, we introduce a method consisting variations machine, deep, transfer learning approaches effective brain tumor (i.e., glioma) segmentation on...

Finansal varlıkların gelecekteki değerlerinin tahmini yatırımcılar için varlıklarını korumak adına önemlidir. 2008 yılında hayatımıza giren ve finansal varlıklar konusunda radikal bir değişiklik olan Bitcoin ise eski yeni yatırımcıların ilgisini çekmiş durumdadır. Ancak Bitcoin, doğası gereği diğer varlıklara göre değerini belirleyen farklı parametreler içermektedir geleneksel tahmin yöntemleri gibi çok hareketli değerlere sahip varlıkları etmekte güçlük çekmektedir. Bu çalışmada değişkenli...

Ant Colony Optimization (ACO) Algorithms have been used to solve many optimization problems in various fields and several algorithms proposed based on ACO metaheuristic the literature. This paper proposes a simple crossover mechanism mean of pheromone tables for algorithms. Main purpose operation is produce solutions or individuals having greater performance than their parents by selecting useful parts. Original don't crossover. Method that we developed employs more one ant colonies also...