Detection of Foreign Materials in Wheat Kernels Using Regional Colour Descriptors

Present paper reports the development of an automated machine vision system for detection of foreign materials in wheat kernels using regional color descriptors. The said system was executed in the form of an integrated flowing pipeline after having proper choice of different possible alternatives at different stages of image processing. A new type of surface colour descriptor is also proposed in this work to define wheat kernel uniquely. The fifteen-element colour descriptor is executed after having thorough comparison of six different colour spaces, each having 72 separate quantifiable components. The fifteen elements of the proposed colour-descriptor, extracted from each segmented region of the sample image, are concatenated in the form of an input to the neural classifier. The neural classifier is trained with Levenberg-Marquardt (LM) learning algorithm to achieve extremely fast convergence. The recognition rate of the executed classifier is found to be more than 99.2% for detection of impurity in unconnected wheat kernels. The results of present investigations are quite promising. The proposed pipeline has potential future in the field of machine vision based quality inspection of wheat and other cereal grains.

Neural Network Classifier of Oil Pollution on the Water Surface when Processing Radar Images

The paper proposes a solution to the problem of detecting oil pollution on a monochrome radar image. The detection of oil pollution in the image includes the solution of three tasks: detecting a dark object on the image, highlighting the main characteristics of a dark object, classifying a dark object as oil pollution or natural slick. Various characteristics of a dark object are proposed based on the contrast between the object and the background. It is proposed to use a neural network as a classifier. The input parameters of the neural network classifier of the dark image object are proposed. A technique for determining the structure of a neural classifier is presented. An algorithm for testing the selected structure of the neural network for the suitability of classifying the dark area on the image of the water surface as oil pollution or wind slick is proposed. The results of the work of the neural network classifier program for detecting abnormal objects in radar images are demonstrated.

Early diagnosis of Lung Cancer with Probability of Malignancy Calculation and Automatic Segmentation of Lung CT scan Images

Computer aided detection system was developed to identify the pulmonary nodules to diagnose the cancer cells. Main aim of this research enables an automated image analysis and malignancy calculation through data and CPU infrastructure. Our proposed algorithm has improvement filter to enhance the imported images and for nodule selection and neural classifier for false reduction. The proposed model is experimented in both internal and external nodules and the obtained results are shown as response characteristics curves.