Scientific community accepted our novel decomposition method based on convolution kernel compensation (CKC). The editorial in Medical and Biological Engineering & Computing says: "A novel approach for decomposition is presented in the article 'Correlation-Based Decomposition of Surface Electromyograms at Low Contraction Forces' by A. Holobar and D. Zazula, and represents a major breakthrough in EMG processing." A complete background and exhaustive experiments were published in IEEE trans. signal process., Sep. 2007, vol. 55, iss. 9, pp. 4487-4496 (COBISS.SI-ID 11597846).
COBISS.SI-ID: 8986134
We discuss domain analysis tools and language development systems that may help to speed up DSL development. The paper was published in the ACM Computing Surveys journal, which is ranked 1st among all computer science. According to Communications of the ACM, Vol. 49, No. 9, September 2006, this paper was among top 10 most popular journal and computing survey articles downloaded in June 2006. It is already used as standard teaching material (e.g., http://www.diku.dk/undervisning/2006f/213/). The paper has been cited 33 times by WoS, 34 times by ACM Digital Library and 200 times by Google.
COBISS.SI-ID: 10259222
With this novel technique we obtain a smooth interpolation between the density values defined in the implicit graphical primitives for rendering clouds without the undesired increase of the field value in areas where the primitives intersect. The method works correctly even in cases when an implicit primitive overlaps completely with another one, which gives us greater flexibility for their positioning. The method disadvantage that weights of all implicit primitives have to be equal can be eliminated by the procedural cloud model with locally determined parameters.
COBISS.SI-ID: 9395478
We describe an efficient technique for adapting control parameter settings associated with differential evolution (DE). This new paradigm was published by the journal which is rank 2 in the artificial intelligence group. The experimental results show that our algorithm with self-adaptive control parameter settings is better than, or at least comparable to, the standard DE algorithm and evolutionary algorithms from literature when considering the quality of the solutions obtained. The proposed method may have great influence on a wide range of computing applications.
COBISS.SI-ID: 10376982
We present an original algorithm for solving constrained Delaunay triangulation based on the sweeping strategy. The algorithm forms new triangles on the boundary of triangulation, called an advancing front. Special heuristics are employed for smoothing the front. Edges are inserted after they are entirely swept. The algorithm is fast and memory friendly.
COBISS.SI-ID: 12266774