Classification of Brain Tumors Using Hybridized Convolutional Neural Network in Brain MRI images

In this research article, we have proposed a novel technique to operate on the Magnetic Resonance Imaging (MRI) data images which can be classified as image classification, segmentation and image denoising. With the efficient utilization of MRI images the medical experts are able to identify the medical disorders such as tumors which are correspondent to the brain. The prime agenda of the study is to organize brain into healthy and brain with tumor in brain with the test MRI data as considered. The MRI based technique is an methodology to study brain tumor based information for the better detailing of the internal body images when compared to other technique such as Computed Tomography (CT).Initially the MRI image is denoised using Anisotropic diffusion filter, then MRI image is segmented using Morphological operations, to classify the images for the disorder CNN based hybrid technique is incorporated, which is associated with five different set of layers with the pairing of pooling and convolution layers for the comparatively improved performance than other existing technique. The considered data base for the designed model is a publicly available and tested KAGGLE database for the brain MRI images which has resulted in the accuracy of 88.1%.

Identification and Classification of Tomato Leaf Diseases Using Machine Learning Techniques

Tomato is cultivated in all countries of the world in fields, glasshouses etc. China, India, USA, Turkey, Egypt, Iran, Italy, Spain and Brazil are the important countries which are cultivating tomatoes. It is most commonly and widely cultivated in India. India is one of the countries in harvesting tomatoes. Tomato is a vital vegetable yield with respect to both income and food. Tomatoes are for the most part summer crops, yet it tends to improve steadily. Naturally, it contains A and C of vitamins which also acts as an antioxidant to prevent cancerous cells. Since the organic product contains novel features, the demand remains the same. A significant and unique feature with high nutrients gains the importance in tomatoes cultivation. Challenges towards the cultivation of tomato made us to plan for an automated machine to detect infection and to increase the productivity. This system automatically detects the infected parts and classify the types of disease which occur on the leaf like early blight, bacterial wilt, Leaf Spot, tomato mosaic virus, septoria leaf spot, leaf curl virus, and tomato spotted wilt disease using gradient anisotropic diffusion filter for pre-processing and then features are extracted using GLCM from the pre-processed leaf

A Brief Survey of Recent Edge-Preserving Smoothing Algorithms on Digital Images

Edge preserving filters preserve the edges and its information while blurring an image. In other words they are used to smooth an image, while reducing the edge blurring effects across the edge like halos, phantom etc. They are nonlinear in nature. Exam?ples are bilateral filter, anisotropic diffusion filter, guided filter, trilateral filter etc. Hence these family of filters are very useful in reducing the noise in an image making it very demanding in computer vision and computational photography applications like de?noising, video abstraction, demosaicing, optical-flow estimation, stereo matching, tone mapping, style transfer, relighting etc. This paper provides a concrete introduction to edge preserving filters starting from the heat diffusion equation in olden to recent eras, an overview of its numerous applications, as well as mathematical analysis, various efficient and optimized ways of implementation and their interrelationships, keeping focus on preserving the boundaries, spikes and canyons in presence of noise. Furthermore it provides a realistic notion for efficient implementation with a research scope for hardware realization for further acceleration.

Frequency division denoising algorithm based on VIF adaptive 2D-VMD ultrasound image

Ultrasound imaging has developed into an indispensable imaging technology in medical diagnosis and treatment applications due to its unique advantages, such as safety, affordability, and convenience. With the development of data information acquisition technology, ultrasound imaging is increasingly susceptible to speckle noise, which leads to defects, such as low resolution, poor contrast, spots, and shadows, which affect the accuracy of physician analysis and diagnosis. To solve this problem, we proposed a frequency division denoising algorithm combining transform domain and spatial domain. First, the ultrasound image was decomposed into a series of sub-modal images using 2D variational mode decomposition (2D-VMD), and adaptively determined 2D-VMD parameter K value based on visual information fidelity (VIF) criterion. Then, an anisotropic diffusion filter was used to denoise low-frequency sub-modal images, and a 3D block matching algorithm (BM3D) was used to reduce noise for high-frequency images with high noise. Finally, each sub-modal image was reconstructed after processing to obtain the denoised ultrasound image. In the comparative experiments of synthetic, simulation, and real images, the performance of this method was quantitatively evaluated. Various results show that the ability of this algorithm in denoising and maintaining structural details is significantly better than that of other algorithms.

Weighted area constraints-based breast lesion segmentation in ultrasound image analysis

<p style='text-indent:20px;'>Breast ultrasound segmentation is a challenging task in practice due to speckle noise, low contrast and blurry boundaries. Although numerous methods have been developed to solve this problem, most of them can not produce a satisfying result due to uncertainty of the segmented region without specialized domain knowledge. In this paper, we propose a novel breast ultrasound image segmentation method that incorporates weighted area constraints using level set representations. Specifically, we first use speckle reducing anisotropic diffusion filter to suppress speckle noise, and apply the Grabcut on them to provide an initial segmentation result. In order to refine the resulting image mask, we propose a weighted area constraints-based level set formulation (WACLSF) to extract a more accurate tumor boundary. The major contribution of this paper is the introduction of a simple nonlinear constraint for the regularization of probability scores from a classifier, which can speed up the motion of zero level set to move to a desired boundary. Comparisons with other state-of-the-art methods, such as FCN-AlexNet and U-Net, show the advantages of our proposed WACLSF-based strategy in terms of visual view and accuracy.</p>

Characterisation of diamond abrasive grains of grinding tools using industrial X-ray computed tomography

In this paper, a characterisation of diamond abrasive grains of grinding tools using industrial X-ray computed tomography (XCT) is carried out. One of the most challenge tasks in the characterisation is extracting the diamond abrasive grains from the XCT volume data. Methods that are able to extract the grains are then developed and introduced in this paper. The first step is to create a triangular mesh surface from the reconstructed volume file using a gradient anisotropic diffusion filter. The second step is to convert the measured greyscale volume into a signed distance field using a global threshold value and then a localised method for grain segmentation. To validate the proposed method, three different types of grinding tool specimens are measured and analysed. Each abrasive grain is segmented and the distributions of grains (with both random and designed patterns) are then calculated, plotted and analysed. The quantitative analysis clearly shows the deviations between the measured distribution and the designed pattern of the grinding tool, which indicates that the proposed method can provide an accurate and comprehensive characterisation of the grinding tools.