Julian Balcerek

Brightness Correction and Stereovision Impression Based Methods of Perceived Quality Improvement of CCTV Video Sequences

In this paper brightness correction and stereovision impression based methods of the perceived quality improvement of CCTV video sequences are presented. These methods are helpful for the monitoring operator in order to evoke attention during a long time observation. Clearness of picture is increased by using local brightness correction method. Another available option is a 3D visualization that can be used, e.g., to observe important image regions, which require an increased attention. Stereovision impression experiments and a real-time 2D to 3D video conversion tool are described.

Mobile applications for driver and pedestrian assistance

In this paper two applications, which fit into the current trend of the development of mobile devices, are described. The first is designed for the driver and the second for the pedestrian assistance. Two traffic events like traffic lights change from the red to the green and vehicle approaching to pedestrian are detected. Proper warnings are generated and the suitable reaction is performed. The car may start movement without unnecessary delay and the pedestrian may change the movement trajectory in order to avoid an accident. Thus, the proper traffic flow and urban safety improvement are supported. Results of initial experiments show a high efficiency of proposed mobile solutions.

Multicriteria Metadata Mechanisms for Fast and Reliable Searching of People Using Databases with Unreliable Records

In this paper an approach to multicriteria search for people is presented using data extracted from telephone calls to emergency services. In the considered application a procedure of searching the most similar object (objects) to the reference object using metadata mechanisms is presented. The proposed solution computes as much reliable result as possible using even unreliable records in the database. In the considered case the commonly used mechanism based on exact matching does not give the best result. Instead, the multicriteria metadata matching mechanisms exploiting weights, probabilities, distances, and correlations among database records are more reliable.

Classification of emergency phone conversations with artificial neural network

This paper presents a series of experiments on the classification of emergency phone conversation records using artificial neural networks (ANNs). Input data which were processed by ANNs were the features of callers and events taken from emergency phone calls. The authors analyzed four variants of classification: the groups of callers which have specified features, the groups of events which have specified features, selected callers, and selected events. Then the efficiency of classification by the ANN (artificial neural network) with various sets of features was compared. Results show that ANNs can properly classify precisely defined callers or events, from e.g. so called ‘black-list’ of callers.

Improvements in upscaling of pixel art

In this paper two approaches to upscaling of low resolution bitmaps created using pixel art digital form are proposed. The methods are aimed to deal with limitations of classical and dedicated algorithms typically used in gaming applications. The proposed spatial methods, proximity-based coefficient correction and transition area restriction, lead to no limitations in scale factor in comparison to the existing pixel-art algorithms. Moreover, in contrast to the classical methods the proposed approaches allow to obtain undistorted and sharp resulting images.

HDR tonal mapping algorithm for mobile devices

In this paper the application for generation of HDR image based on two consecutive images (underexposed and overexposed) for Android mobile operating system is presented. The implemented software preserves a lot of image details and maintains a low execution time. These features are particularly important for pictures taken using mobile devices in emergency situations. Such photos may constitute evidence that a threat occurred, was properly recognized, or someone committed a crime. HDR images can be also used in mobile systems for supporting pedestrians or drivers. Obtained results indicate on a high effectiveness of the presented solution.

Approach to local contrast enhancement

In this article a simple and fully automatic approach to local contrast enhancement is presented, maintaining reasonable computational complexity and feasibility of simultaneous calculations on multiple processors. The authors propose an algorithm, which allows to obtain images with artificially increased dynamic range. The resulting images do not contain unnatural artifacts and are close to images perceived by humans. Results of experiments show that the described method offers very good results even for images obtained with a large range of differences in the exposures.

Speaker identification - oriented multimodal database system model

In the paper, the multimodal database system model designed to store the audio, video, graphical and text file records for speaker identification is proposed. The opportunity to make the records accessible on-line via Web in the client-server architecture using an open source DBMS is provided, and the access involves sharing, loading, modifying or completing the records depending on the users authorization level. Due to the modular structure and the categorization concept, the scalability of the system along the macro axis (purpose-oriented component databases) and micro axes (number of speakers/recording conditions, language, etc.) has been obtained. A specific solution has been proposed regarding a need for dealing with great volume files.