Standard test image

About: Standard test image is a research topic. Over the lifetime, 5217 publications have been published within this topic receiving 98486 citations.

A Self-Supervised Decision Fusion Framework for Building Detection

Abstract: In this study, a new building detection framework for monocular satellite images, called self-supervised decision fusion (SSDF) is proposed. The model is based on the idea of self-supervision, which aims to generate training data automatically from each individual test image, without human interaction. This approach allows us to use the advantages of the supervised classifiers in a fully automated framework. We combine our previous supervised and unsupervised building detection frameworks to suggest a self-supervised learning architecture. Hence, we borrow the major strength of the unsupervised approach to obtain one of the most important clues, the relation of a building, and its cast shadow. This important information is, then, used in order to satisfy the requirement of training sample selection. Finally, an ensemble learning algorithm, called fuzzy stacked generalization (FSG), fuses a set of supervised classifiers trained on the automatically generated dataset with various shape, color, and texture features. We assessed the building detection performance of the proposed approach over 19 test sites and compare our results with the state of the art algorithms. Our experiments show that the supervised building detection method requires more than 30% of the ground truth (GT) training data to reach the performance of the proposed SSDF method. Furthermore, the SSDF method increases the F-score by 2 percentage points (p.p.) on the average compared to performance of the unsupervised method.

Lossless Microarray Image Compression using Region Based Predictors

Abstract: Microarray image technology is a powerful tool for monitoring the expression of thousands of genes simultaneously. Each microarray experiment produces large amount of image data, hence efficient compression routines that exploit microarray image structures are required. In this paper we introduce a lossless image compression method which segments the pixels of the image into three categories of background, foreground, and spot edges. The segmentation is performed by finding a threshold value which minimizes the weighted sum of the standard deviations of the foreground and background pixels. Each segment of the image is compressed using a separate predictor. The results of the implementation of the method show its superiority compared to the well-known microarray compression schemes as well as to the general lossless image compression standards.

System and method for performing wavelet-based texture feature extraction and classification

Abstract: A system and method are disclosed for performing wavelet-based local texture feature extraction and classification procedures. Image data is initially provided to include a query image and a series of test images. A feature detector calculates image parameters corresponding to the image data. The image parameters include mean absolute values, variance values, and texture angles. The feature detector utilizes the image parameters to calculate distance values that represent texture similarity characteristics between the query image and each of the test images. The feature detector then evaluates the distance values to determine one or more matching images from among the test images.

Method for producing super-resolution images and nonlinear digital filter for implementing same

Abstract: The invention relates to the field of photo and video images and can be used for producing high-quality images of visually close objects using a camera or video camera equipped with sensors with an electronic shutter. The technical result consists in increasing the resolution of an image when a plurality of frames of low resolution are used for producing one frame of high resolution (i.e. super-resolution), as well as the possibility of high-speed capture of a plurality of frames of an image while scanning only part of the sensor. The result is achieved in that a plurality of frames is exposed, initial images are produced by means of a reading from the sensor in the form of a continuous sequence of frames with high-speed capture, during which the frequency of the frames is inversely proportional to the magnitude of that part of the light-sensitive region of the sensor which is being scanned, said initial images are aligned, an enhanced image is produced and this image is filtered using a nonlinear filter, which comprises a neural network which is pretrained using a test image comprising radial and sinusoidal test charts, as well as reference points. Furthermore, the filtration comprises supplying premodified digitized data to the neural network, wherein the modification of said data comprises: isolating the low-frequency component, arranging the pixels element-by-element, reading the low-frequency component from the arranged pixels, and subsequently standardizing said pixels. Then, the data at the output of the neural network are subjected to inverse standardization, and the low-frequency component is added to the value at the output of the neural network.