Multi-Lingual Character Recognition and Extraction using Recurrent Neural Networks
A Novel Approach to Smart Autonomous Monitoring of Indoor Plant Health Based on Leaf Color
Edge Preserving Image Denoising with Deep Networks and Ensemble Convolutional Neural Network Architectures
Neural Network Approach to Detect Renal Calculi
Enhancing Fabric Quality Control: Implementing Real-Time Defect Detection with Image Processing Techniques and Arduino
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
Sickle cells are abnormalities of the Red Blood Cells (RBCs). The focus of this research work is to detect sickle cells, normal RBCs, and to separate the overlapping cells present in a blood film image and also to identify the percentage of each detected cells. The image processing techniques used for the detection of these RBCs consist of five main steps which are, image acquisition and reading, image preprocessing, feature extraction, RBCs classification, and final result. The Marker-Based Watershed Segmentation (MBWS) method was used to separate the detected overlapping cells. To test the accuracy of this system, the system's result was compared to the result of the traditional method of detection and count of RBCs used in the hospital. The identification and separation of the overlapping cells performed by the proposed system made the system provide more accurate result as compared to the traditional method. The limitations encountered by the system includes, inadequate separation of the overlapping cells due to the under segmentation problem of the technique (Marker-Based Watershed Segmentation) used, the quality of images used, and also the cells at the boundary were eliminated together with its diagnostic features which it may possess.
The detection of Multiple Sclerosis (MS) brain tissue is essential for several neuroimaging studies. In this paper, we implement a Particle Region Growing Algorithm (PRGA) in order to segment brain tissue affected by the MS. To concentrate the brain, White Matter (WM) lesions are required more attention to find out the abnormal brain tissue along with normal brain tissue in T2W MRI brain image scan. However, the sensitivity and specificity of MS lesion detection and segmentation with the different method approaches have been inadequate. In this paper, we concentrate on the White Matter (WM) and Gray Matter (GM) of MS lesion affected the T2 weighted transverse view of the brain MR image. We carried out extensive experiments with MS patients MRI image data. In this work, we found a new approach that leads to a substantial improvement in the sensitivity and specificity of MS lesion detection by using a PRGA segmentation algorithm.
Today in the agricultural and food industry, a method of quality evaluation is required in order to grade the fruits and vegetables, to process them in huge scale. This would result in the increase of profits and prompted supply to the customers. So a quality evaluation method is required to handle the supply chain management. The proposed method is based on image processing technique for automated grading of fruits and vegetables. The method considers according to the maturity level in terms of quality attributes, such as size, shape, and surface defect. In this system, different image processing techniques like pre-processing of image, features extraction frame extraction are being implemented for final gradation of fruits and vegetables. Different attributes present in the sample will be analyzed. In this way, the proposed system would grade the given vegetables and fruits based on experts’ perception by using different image processing techniques.
In deep learning neural network architectures, the term “deep” usually refers to the number of hidden layers in the neural network. Traditional neural networks only contain 2-3 hidden layers, while deep networks can have as many as 150. One of the most popular types of deep learning neural networks is known as Convolution Neural Networks (CNN). Image Analysis is the process to analyse various feature of the image such as segmentation of the image, classification of the images, etc. In this proposed work, a CNN based Image Analysis System has been proposed to detect the objects in real time image and classify the image based on the feature of extraction. The proposed method is implemented by using MATLAB and Python software. The system is trained with the CIFAR -10 and MINST Dataset and own image database. The performance of the proposed method is evaluated by using accuracy and loss factor. The proposed system is tested with unknown samples of images and also trained for accuracy and average test loss.
Medical images such as Magnetic Resonance Images (MRI) and Computed Tomography (CT) are the two most emerging imaging technologies. Processing these are useful in identification of disease and also in diagnosing a specific organ. In this paper, we have proposed a procedure for the detection of brain tumour from MRI images of brain using Image Processing techniques such as filtering in Frequency Domain, Thresholding, and Morphological Reconstruction. These techniques are implemented on MRI images using MATLAB Image Processing toolbox. Image is converted into Frequency Domain and Butterworth High pass filters are applied on the image and then opening-closing by reconstruction are applied on the image to detect the tumour present in the image and also the area of the tumour is identified.
White Blood Cell (WBC) is known as leukocyte or white corpuscle. It defends the body against infection and disease by removing foreign materials and cellular debris. The normal range of WBC in healthy people varies from 4,000 to 11,000 WBCs per microliter. If this count lowers, it indicates various diseases, such as viral infection, cancer, leukemia, etc. and high count indicate diseases such as infection, stress, allergies, etc. Traditional manual inspection of blood smear image is replaced by computerized technology and currently with the help of deep learning approach, we can extract accurate count and detailed condition of WBCs, which helps pathologist in terms of less labour and less time. We use WBC dataset having 12,500 images for experimentation on deep convolutional neural networks like AlexNet, VGG16, VGG19, InceptionV3, and ResNet50. To get better result, each network is modified by changing the layers in original architecture. Further various hyper parameters, such as learning rate, regularization, training epochs, batch sizes, etc., are adjusted to improve accuracy. All these networks have been compared for selection of appropriate network. InceptionV3 gives highest accuracy (99.6%) for WBC classification.
