Diagnosis of Diabetic Retinopathy through Retinal Fundus Images and 3D Convolutional Neural Networks with Limited Number of Samples

Diabetic retinopathy (DR) is a worldwide problem associated with the human retina. It leads to minor and major blindness and is more prevalent among adults. Automated screening saves time of medical care specialists. In this work, we have used different deep learning (DL) based 3D convolutional neural network (3D-CNN) architectures for binary and multiclass (5 classes) classification of DR. We have considered mild, moderate, no, proliferate, and severe DR categories. We have deployed two artificial data augmentation/enhancement methods: random weak Gaussian blurring and random shift along with their combination to accomplish these tasks in the spatial domain. In the binary classification case, we have found the performance of 3D-CNN architecture trained by deploying combined augmentation methods to be the best, while in the multiclass case, the performance of model trained without augmentation is the best. It is observed that the DL algorithms working with large volumes of data may achieve better performances as compared to the methods working with small volumes of data.

UnCanny: Exploiting Reversed Edge Detection as a Basis for Object Tracking in Video

Few object detection methods exist which can resolve small objects (<20 pixels) from complex static backgrounds without significant computational expense. A framework capable of meeting these needs which reverses the steps in classic edge detection methods using the Canny filter for edge detection is presented here. Sample images taken from sequential frames of video footage were processed by subtraction, thresholding, Sobel edge detection, Gaussian blurring, and Zhang–Suen edge thinning to identify objects which have moved between the two frames. The results of this method show distinct contours applicable to object tracking algorithms with minimal “false positive” noise. This framework may be used with other edge detection methods to produce robust, low-overhead object tracking methods.

Assessing Deep Convolutional Neural Networks and Assisted Machine Perception for Urban Mapping

High-spatial-resolution satellite imagery has been widely applied for detailed urban mapping. Recently, deep convolutional neural networks (DCNNs) have shown promise in certain remote sensing applications, but they are still relatively new techniques for general urban mapping. This study examines the use of two DCNNs (U-Net and VGG16) to provide an automatic schema to support high-resolution mapping of buildings, road/open built-up, and vegetation cover. Using WorldView-2 imagery as input, we first applied an established OBIA method to characterize major urban land cover classes. An OBIA-derived urban map was then divided into a training and testing region to evaluate the DCNNs’ performance. For U-Net mapping, we were particularly interested in how sample size or the number of image tiles affect mapping accuracy. U-Net generated cross-validation accuracies ranging from 40.5 to 95.2% for training sample sizes from 32 to 4096 image tiles (each tile was 256 by 256 pixels). A per-pixel accuracy assessment led to 87.8 percent overall accuracy for the testing region, suggesting U-Net’s good generalization capabilities. For the VGG16 mapping, we proposed an object-based framing paradigm that retains spatial information and assists machine perception through Gaussian blurring. Gaussian blurring was used as a pre-processing step to enhance the contrast between objects of interest and background (contextual) information. Combined with the pre-trained VGG16 and transfer learning, this analytical approach generated a 77.3 percent overall accuracy for per-object assessment. The mapping accuracy could be further improved given more robust segmentation algorithms and better quantity/quality of training samples. Our study shows significant promise for DCNN implementation for urban mapping and our approach can transfer to a number of other remote sensing applications.

What regions of Chinese characters are crucial for recognition? A web-based study

The study involves approximately 250 Chinese university students from eight institutions to determine what parts of a representative sample of Chinese characters are crucial to their correct identification. A web-based experimental platform was used to present 102 simplified characters to participants. The characters were partially obscured using a Gaussian blurring technique. The direction of maximum blur could either be from top to bottom, bottom to top, left to right, or right to left. Participants were asked to identify the blurred character and type its pinyin. Overall, participants correctly identified 88% of characters. The effects of all forms of blurring on correct recognition were correlated with character structures. Phonetic radicals seem to be more sensitive to the blurring than semantic radicals, while the radical transparency and radical frequency also play a role in the recognition accuracy. The blurring conditions that impacted most significantly on correct recognition were top to bottom and bottom to top, which caused, respectively, the upper and lower parts of the character to be obscured.

Simulation of a muographic analysis of a volcanic dome in Geant4

Here we present a Monte Carlo simulation of a muographic campaign on Methana volcano, Greece. In order to estimate the absorption parameters and the pattern of muon scattering at various incident energies (GeV to TeV), a radar-derived digital terrain model (DTM) was submitted to irradiation by horizontal muons in Geant4 and the penetrating muons were collected by a hypothetical MicroMegas particle detector on the other side of the DTM. Monte Carlo simulation demonstrated that muon energies at least as high as 10 TeV are required for whole-scale radiography of Methana and one has to reduce the scale of study to smaller structures (e.g. ~ 600 m - wide volcanic domes) in order to exploit the more affluent lower energy muons (~ 600 GeV). Coulomb scattering, on the other hand, brings about deflection of muon trajectories away from the detector, resulting in loss of information. Additionally, scattering adds Gaussian blurring to the scanned objects. With the intention of improving contrast and extract objects in muographic image we recommend the use of spatial operators (filters) employed in image analysis.

THE SYSTEM RECOGNIZES SURFACE DEFECTS OF MARBLE SLABS BASED ON SEGMENTATION METHODS

A system for recognizing surface defects in marble slabs is proposed. The pattern recognition method based on segmentation methods was further developed. The algorithm of the recognition system. The article describes methods for determining damage from digital images on various hard surfaces. Research in this field is relevant for a wide range of industrial enterprises that specialize in the production of various kinds of materials: parts, marble slabs, building materials, etc. To solve this problem, it is proposed to use the k-means clustering method. It has been experimentally established that Gaussian blurring algorithms, the Hough transform, and the Kenny algorithm are best suited for recognizing defects on the surface of a marble slab. The developed complex method based on the theory of pattern recognition allows you to quickly identify defects and damage on the surfaces of marble slabs. On the basis of the method, a system for understanding defects is implemented in software. The main stages of the system are described in the article. The results of the analysis of the image of the surface of the marble slab on a specific example are presented. The developed complex method based on the theory of pattern recognition allows you to quickly identify defects and damage on the surfaces of marble slabs. On the basis of the method, a system for understanding defects is implemented in software. The main stages of the system are described in the article. The results of the analysis of the image of the surface of the marble slab on a specific example are presented.

Detection of linear features including bone and skin areas in ultrasound images of joints

Identifying the separate parts in ultrasound images such as bone and skin plays a crucial role in the synovitis detection task. This paper presents a detector of bone and skin regions in the form of a classifier which is trained on a set of annotated images. Selected regions have labels: skin or bone or none. Feature vectors used by the classifier are assigned to image pixels as a result of passing the image through the bank of linear and nonlinear filters. The filters include Gaussian blurring filter, its first and second order derivatives, Laplacian as well as positive and negative threshold operations applied to the filtered images. We compared multiple supervised learning classifiers including Naive Bayes, k-Nearest Neighbour, Decision Trees, Random Forest, AdaBoost and Support Vector Machines (SVM) with various kernels, using four classification performance scores and computation time. The Random Forest classifier was selected for the final use, as it gives the best overall evaluation results.

Detection of linear features including bone and skin areas in ultrasound images of joints

Identifying the separate parts in ultrasound images such as bone and skin plays the crucial role in synovitis detection task. This paper presents a detector of bone and skin regions in the form of a classifier which is trained on a set of annotated images. Selected regions have labels: skin or bone or none. Feature vectors used by the classifier are assigned to image pixels as a result of passing the image through the bank of linear and nonlinear filters. The filters include Gaussian blurring filter, its first and second order derivatives, Laplacian as well as positive and negative threshold operations applied to the filtered images. We compared multiple supervised learning classifiers including Naive Bayes, k-Nearest Neighbour, Decision Trees, Random Forest, AdaBoost and Support Vector Machines (SVM) with various kernels, using four classification performance scores and computation time. The Random Forest classifier was selected for the final use, as it gives the best overall evaluation results.

Detection of linear features including bone and skin areas in ultrasound images of joints

Identifying the separate parts in ultrasound images such as bone and skin plays the crucial role in synovitis detection task. This paper presents a detector of bone and skin regions in the form of a classifier which is trained on a set of annotated images. Selected regions have labels: skin or bone or none. Feature vectors used by the classifier are assigned to image pixels as a result of passing the image through the bank of linear and nonlinear filters. The filters include Gaussian blurring filter, its first and second order derivatives, Laplacian as well as positive and negative threshold operations applied to the filtered images. We compared multiple supervised learning classifiers including Naive Bayes, k-Nearest Neighbour, Decision Trees, Random Forest, AdaBoost and Support Vector Machines (SVM) with various kernels, using four classification performance scores and computation time. The Random Forest classifier was selected for the final use, as it gives the best overall evaluation results.

Perceptual Hashing-Based Image Copy-Move Forgery Detection

This paper proposes a blind authentication scheme to identify duplicated regions for copy-move forgery based on perceptual hashing and package clustering algorithms. For all fixed-size image blocks in suspicious images, discrete cosine transform (DCT) is used to obtain their DCT coefficient matrixes. Their perceptual hash matrixes and perceptual hash feature vectors are orderly addressed. Moreover, a package clustering algorithm is proposed to replace traditional lexicographic order algorithms for improving the detection precision. Similar blocks can be identified by matching the perceptual hash feature vectors in each package and its adjacent package. The experimental results show that the proposed scheme can locate irregular tampered regions and multiple duplicated regions in suspicious images although they are distorted by some hybrid trace hiding operations, such as adding white Gaussian noise and Gaussian blurring, adjusting contrast ratio, luminance, and hue, and their hybrid operations.