Android platform due to open source characteristic and Google backing has the largest global market share. Being world's most popular operating system, it drawn attention of cyber criminals particularly through wide distribution malicious applications. This paper proposes an effectual machine-learning based approach for Malware Detection making use evolutionary Genetic algorithm discriminatory feature selection. Selected features from are used train machine learning classifiers their...

Object detection in ultrasound images is difficult problem mainly because of relatively low signal-to-noise ratio. This paper deals with object the noisy using modified version Viola-Jones detector. The method describes carotid artery longitudinal section B-mode images. detector primarily trained by AdaBoost algorithm and uses a cascade Haar-like features as classifier. main contribution this for section. creates classifiers automatically genetic algorithms. We also created post-processing...

The proposed work presents a zero watermarking method for to solve the issue of medical image security telemedicine, tele-radiology & tele-opthalmology applications. This provides tele-medicine application without tempering and no loss clinical information. Local features in Singular value decomposition (SVD) domain are used generate digital binary code (Master Share) each fundus image. master share is strategically combined with encrypted patient ID resulting into secret share. At diagnosis...

Abstract In this work, we proposed a new data acquisition approach that significantly improves the repetition rates of Laser-Induced Breakdown Spectroscopy (LIBS) experiments, where high-end echelle spectrometers and intensified detectors are commonly used. The moderate recent LIBS systems caused by utilization their slow full frame ( i.e . echellogram) readout speeds with consequent necessity for echellogram-to-1D spectrum conversion (intensity vs wavelength). Therefore, investigated...

Graphic Processing Units (GPUs) offer cheap and high-performance computation capabilities by offloading compute-intensive portions of the application to GPU, while remainder code still runs on a CPU. This paper introduces an multi-GPU CUDA implementation training object detection using Viola-Jones algorithm that has accelerated two most time consuming operations in process dual-core NVIDIA GeForce GTX 690. When compared single thread Intel Core i7 3770 with 3.7 GHz frequency, first part was...

Magnetic resonance imaging (MRI) is a visualizing method used in radiology that enables viewing internal structures of the body. Using several mathematical methods with data retrieved from MRI it possible to quantify brain compartment volume, which has many applications cognitive, clinical and comparative neurosciences. This paper introduces new fully automatic method, can measure volume tissue using scans obtained devices. The introduced this was trained on taken 12 patients result...

The paper proposes an image processing algorithm to grade the severity of Non Proliferative Diabetic Retinopathy. For this disease most important parameter classify stage is proximity abnormalities from centre Macula. proposed provides efficient grading technique by segmenting fundus into specific regions interest and avoids redundancy in computation. Instead detecting for whole image, method emphasizes on segmented abnormalities, thereby reducing computation time significantly. Furthermore,...

Using modern Graphic Processing Units (GPUs) becomes very useful for computing complex and time consuming processes. GPUs provide high-performance computation capabilities with a good price. This paper deals multi-GPU OpenCL implementation of k-Nearest Neighbor (k-NN) algorithm. The proposed algorithm achieves acceleration up to 750x in comparison single thread CPU version. common k-NN was modified be faster when the lower number k neighbors is set. performance verified two dual-core NVIDIA...

This paper deals with localization of Temporomandibular Joint Disc (TJD) in Magnetic Resonance Images (MRI). Since the contrast TJD is quite low when compared to noise ratio displayed using MRI, its detection complicated. Therefore method described this are not focused disk itself but detect most significant objects around TJD, which has usually much higher contrast. For automatic assessment, a training set containing 160 samples (80 positive and 80 negative) were created published several...

Body auscultation is an easy and non-invasive method for detection of diseases in human body. The conventional a bit time consuming requires professionals diagnosis. Automatic diagnosis using heart sound can be great help the rural areas where professional not available. proposed work presents automatic efficient classification sound. Mel Frequency Cepstral Coefficient (MFCC) features are extracted from sounds Supervised used to separate normal abnormal diseases. was tested on comprehensive...

Using modern Graphic Processing Units (GPUs) becomes very useful for computing complex and time consuming processes. GPUs provide high–performance computation capabilities with a good price. This paper deals multi–GPU OpenCL CUDA implementations of k–Nearest Neighbor (k–NN) algorithm. work compares performances OpenCLand where each them is suitable different number used attributes. The proposed algorithm achieves acceleration up to 880x in comparison witha single thread CPU version. common...

Acrylamide is a toxic carcinogenic material that commonly occurs in heated starchy food items like potato crisps. Identification of such chemical using conventional laboratory based methods time consuming and expensive may need specialized manpower for testing. This paper proposes an efficient image processing non-destructive testing method identification acrylamide The the crisps automatically segmented features are extracted from images wavelet domain. These then analyzed presence these...

In recent years,many efforts have been devoted to image segmentation. Although for a man general segmentation is considered an easy task, computers it still be difficult, computationally intensive and unresolved task. This work presents innovative algorithm combining theory of artificial intelligence knowledge human eye anatomy. The resulting has not ambitions universal like brain but can trained perform on selected domain. effectiveness the demonstrated examples.

Emotions play big role in our everyday communication and contain important information.This work describes a novel method of automatic emotion recognition from textual data.The is based on well-known data mining techniques, approach parallel run SVM (Support Vector Machine) classifiers, text preprocessing 3 optimization methods: sequential elimination attributes, parameter token groups, extending train sets during practical testing production release final tuning.We outperformed current...

The computer vision connected to image understanding becomes more and important in everyday life. This paper concerns the (dis)similarity assessment grouping. main contribution of this is method for based on dominant colors. experimental results showed better than Direct Pixel Similarity Color Histograms proved be capable detecting images similar target image.

This paper presents a new approach to the classifier design used in Viola-Jones object detector implemented RadpidMiner Image Mining Extension. The proposed this is fact creation of classification tree designed by genetic programming algorithm. resulting as an alternative standard cascade In paper, shown, incorporation into operator described, and experimental results face process are depicted compared

Video scene segmentation is a process for dividing video into semantically meaningful blocks. This can help e.g. search engines to divide better manageable parts and enable more relevant in video. Unfortunately, based on the semantic therefore it difficult task computers. work preliminary study involved supervised segmentation, which driven by way how human segments scenes movie. Since these represent video, be used annotation also searching videos. As training set, only high quality movies...

Scene classification based on global features. It can be used, for example, annotating large databases of photos. The whole process has several steps. first step is features extraction, and then the distance between a new image reference images calculated. A model trained to classify this distance. was created using Naïve Bayes classifier. To improve accuracy forward selection which optimizes group attributes. overall performance testing dataset 69.76%.

This paper introduces an optimization algorithm for job shop scheduling problem in logistic warehouses. The is based on genetic programming and uses parallel processing. For better performance a new method called "priority rules" was proposed. We found out that the three proposed priority rules help to prevent stuck local optima get results from optimization. Algorithm tested with batch of tests data real warehouse synthetic generated randomly (inspired by world scenarios). indicate...

With growing number of stored image data, search and similarity problem become more important.The answer can be solved by Content-Based Image Retrieval systems.This paper deals with an using measures based on circular sectors method.The method is inspired human eye functionality.The main contribution the a modified that increases accuracy for about 8% in comparison original approach.Here proposed has used HSB colour model median function feature extraction.The approach uses RGB mean...

Since security situation in the world is changing, monitoring of protected areas using surveillance systems has been increased significance recent years. Although today object detection methods significantly improved accuracy, for real situations, where video stream basically a low resolution and objects are often small blurry, still struggling with precise detection. The key parts any system 1) person then also 2) recognition, which must perform real-time processing. This paper deals...

Abstract The topic of the presented paper is a promising approach to achieve optimal Design Experiment (DoE), i.e. spreading points within design domain, using simulation discrete dynamical system interacting particles an n-dimensional space. mutually repelling represents physical analogy Audze-Eglājs (AE) optimization criterion and its periodical modification (PAE), respectively. compares performance two approaches implementation: single-thread process JAVA language environment massively...