Text documents screen watermarking by changing background brightness in the interline spacing

One of the most common ways documents leak is taking a picture of document displayed on the screen. For investigation of such cases data leakage prevention technologies including screen watermarking are used. The article gives short review on the problem of screen shooting watermarking and the existing research results. A novel approach for watermarking text images displayed on the screen is proposed. The watermark is embedded as slight changes in luminance into the interline spacing of marked text. The watermark is designed to be invisible for human eye but still able to be detected by digital camera. An algorithm for extraction of watermark from the screen photo is presented. The extraction algorithm doesn’t need the original image of document for successful extraction. The experimental results show that the approach is robust against screen-cam attacks, that means that the watermark stays persistent after the process of taking a photo of document displayed on the screen. A criterion for watermark message extraction accuracy without knowledge about the original message is proposed. The criterion represents the probability that the watermark was extracted correctly.

Simulation and Prototyping of an Orbital Lightning Detector

The problem of detecting lightning flashes from space is becoming more and more important as the remote sensing of Earth, climatology and atmospheric physics develop. Orbital lightning detectors are designed both in Russia and abroad. The paper considers the problem of mathematical and physical simulation of a high-speed camera designed to observe lightning flashes from onboard spacecraft in low Earth orbit. Our previous works substantiated the exterior design of the lightning detector, computed its properties and described the algorithms behind the software. In order to validate the design results, we need to prototype the instrument. We describe the main problems of detecting lightning flashes as observed from space. We show a computer simulation method producing snapshots that takes into account the interference generated by the background and the photodetector. We studied the problems of recreating the target environment during prototyping of the lightning detector, accounting for the properties of the phenomenon observed, that is, the flash spot dimensions and the ratio of the background brightness to that of the lightning. We describe the principle of comparing these two types of snapshots (taken by the prototype and synthesised by the software) and compared the results of processing the images obtained. The results matched, which allowed us to validate the snapshot processing algorithm and confirmed that the lightning detector simulation method developed is correct

Night Sky Background Brightness Estimation by the Example of the St. Petersburg City

Within the big city boundaries, artificial component of the sky glow increases every year. The main reasons are the expansion of the cities, the rapid growth of technical capabilities, inefficient lighting design without master-planning and lack of quality control of lighting projects. The data of astronomical observations confirm a significant brightness increasing in the lower atmosphere due to factors of terrestrial origin. The problem is mostly acute for observatories are located near major cities, which are struggling for the possibility of further research. Night sky background glow estimation is an actual direction for research in the modern world. The paper considers a model for calculating the sky brightness for the St. Petersburg city. According to the developed model observation position is located near the Pulkovo Observatory. The model is based on the Garstang’s method with use of Python programming language.

Binocular Coordination in Reading When Changing Background Brightness

Contradicting results concerning binocular coordination in reading have been reported: Liversedge et al. (2006) reported a dominance of uncrossed fixations, whereas Nuthmann and Kliegl (2009) observed more crossed fixations in reading. Based on both earlier and continuing studies, we conducted a reading experiment involving varying brightness of background and font. Calibration was performed using Gabor patches presented on grey background. During the experimental session, text had to be read either on dark, bright, or grey background. The data corroborates former results that showed a predominance of uncrossed fixations when reading on dark background, as well as those showing a predominance of crossed fixations, when reading on bright background. Besides these systematic shifts, the new results show an increase in unsystematic variability when changing the overall brightness from calibration to test. The origins of the effects need to be clarified in future research.

A multi-layered image format for the web with an adaptive layer selection algorithm

In this paper we present a proposed multi-layered image format for use on the web. The format implements an algorithm for selecting adequate layer images depending on the image container's surroundings and size. The layer selection depends on the weighted average brightness of the underlying web page background within the bounds of the image. The proposed image format supports multiple image layers with adjoined thresholds and activation conditions. Depending on these conditions and the underlying background, a layer's visibility will be adequately set. The selection algorithm takes into account the background brightness, each layer's adjoined threshold values, and other newly introduced layer conditions. <br><br><font color="red"><b> This article has been corrected. Link to the correction <u><a href="http://dx.doi.org/10.2298/SJEE1803371E">10.2298/SJEE1803371E</a><u></b></font>

Reserch on Removal Technology of Snowflakes in Video Images

Using improved K-means clustering method to remove snowflakes in video images getted in snowy day is put forward. According to the brightness changes of the image, we clustering estimate brightness changes of all the pixels in the image by using background brightness values instead of the brightness values of the snowflakes to achieve the goal of removing snowflakes. The results of experiment show that the processing effect of the method is very obvious for light snow and heavy snow, and even storm snow.It can get good results.