NON-INVASIVE NEONATAL GOLDEN HUE DETECTOR
Species Classification and Disease Identification Using Image Processing and Convolutional Neural Networks
A novel meta-heuristic jellyfish Optimize for Detection and Recognition of Text from complex images
Rice Leaf Disease Detection Using Convolutional Neural Network
Comparative Analysis of usage of Machine learning in Image Recognition
Identification of Volcano Hotspots by using Resilient Back Propagation (RBP) Algorithm Via Satellite Images
Data Hiding in Encrypted Compressed Videos for Privacy Information Protection
Improved Video Watermarking using Discrete Cosine Transform
Contrast Enhancement based Brain Tumour MRI Image Segmentation and Detection with Low Power Consumption
Denoising of Images by Wavelets and Contourlets using Bi-Shrink Filter
This study developed a brightness enhancement technique for video frame pixel intensity improvement. Frames extracted from the six sample video data used in this work were stored in the buffer as images. Noise was added to the extracted image frames to vary the intensity of their pixels so that these pixel values of noisy images differ from their true values in order to determine the efficiency of the developed technique. Simulation results of this paper showed improvement in pixel intensity and histogram distribution. The Peak to Signal plus Noise Ratio evaluation showed that the efficiency of the developed technique for both grayscale and coloured video frames were improved by PSNR of 12.45%, 16.32%, 27.57%, and 19.83% over those of the grey level colour (black and white) images for the NAELS1.avi, NAELS2.avi, NTA1.avi, and NTA2.avi, respectively. Also, a percentage improvement of 28.93% and 31.68% were obtained for the coloured images over the grey level images for Akiyo.avi and Forman.avi benchmark video frames, respectively.
Nowadays, the veracity, velocity, values, and size of data are growing exponentially. In fact, the data is growing beyond the capacity of current hardware facilities. This resulted in high cost of storing the data. Perhaps, some of the data stored are not very useful and create problems when mining the data, to make some sense out of it. Feature selection is a step forward towards reducing the unnecessary huge amount of the stored data. In this study, Flower Pollination Algorithm (FPA) along with its binary version (BFPA) are used for feature selection on some medical datasets. The results obtained is in favor of the BFPA with better classification accuracy of over 90% on some of the datasets and fewer number of features compared to FPA, improved harmony search with rough set together with particle swarm optimization with rough set. Hence, the experimental results demonstrate the efficiency and effectiveness of BFPA as the best technique among the evaluated methods for feature selection particularly on medical datasets.
The Micro-calcification which is an early sign of breast cancer is hard to find due to its small size, poor contrast, and blurry image boundary. Pectoral Muscles on mammograms are soft tissues of the body other than breast muscles, which looks like a cancer. The fuzzy algorithms used in this scenario are fuzzy opertors that analyze the image at pixel level to detect abnormalities and identify the location of abnormalities on the breast and pectoral muscles. This paper describes a technique which consist of five steps to find location of cancers in breast by removing pectoral muscles 1) To enhance the quality of poor breast images 2) preconisation of the breast shape 3) extract the cancer part from the breast images 4) removing the Pectoral muscles depends on the orientation of the breasts 5) location of the cancer part on breast images. The result shows the possibility and adequacy of the proposed approach.
The aim of this work is to efficiently detect breast cancer at an early stage and reduce the death rates of women. The purpose of this work is to identify the tumor present in the breast region of mammogram image as benign or malignant as these images are generally of low quality and sometimes radiologists need to seek second opinion to come to the conclusion that cancer is present. The image processing procedure is applied to detect breast cancer from mammographic ROI image. Earlier doctors used MRI, CT-scan, Ultrasound techniques to detect breast cancer, using which it was difficult to identify cancerous tumour at an early stage. The proposed methodology uses mammography technique to identify the tumor present in the breast region. The discrete wavelet and Contourlet transforms are used to decompose the given gray-scale image. The statistical and textual features are being extracted from the coefficients of spatial domain along with frequency domain values. The classification of mammographic ROI image is performed using support vector and artificial neural classifiers. The tool used in this work is Matlab. This work is recommended to study for all those working on breast cancer area of an image processing domain.
Image compression techniques find a wide role in the field of underwater image processing. The wavelet based image compression algorithm performance mainly depends on the encoding methods adopted. In this work, symlet wavelet and hybrid encoding techniques such as Set Partitioning Hierarchy Tree and Huffman encoding are applied on an underwater image and the performance is estimated by compression ratio and PSNR. The results clearly indicate that hybrid encoding performed well. In order to assess its value, PSNR and CR are calculated. The best compression algorithm is chosen based on a compromise between PSNR and CR. The underwater images are compressed in this work.
