Niko Lukač

Chain code lossless compression using move-to-front transform and adaptive run-length encoding

Chain codes are the most size-efficient representations of rasterised binary shapes and contours. This paper considers a new lossless chain code compression method based on move-to-front transform and an adaptive run-length encoding. The former reduces the information entropy of the chain code, whilst the latter compresses the entropy-reduced chain code by coding the repetitions of chain code symbols and their combinations using a variable-length model. In comparison to other state-of-the-art compression methods, the entropy-reduction is highly efficient, and the newly proposed method yields, on average, better compression.

A universal chain code compression method

A novel universal algorithm for various chain code compressions is presented.New chain code binarization scheme is proposed.The compression is based on RLE and variation of LZ77.The method achieves on average higher compression than state-of-the-art methods. This paper introduces a new approach for lossless chain code compression. Firstly, the chain codes are converted into the binary stream, independent on the input chain code. Then, the compression is done using three modes: RLE 0 , LZ 77 0 and COPY. RLE 0 compresses the runs of the 0-bits, LZ 77 0 is a simplified version of LZ 77 and handles the repetitions within the bit stream, whilst COPY is an escape mode used, when the other two methods are unsuccessful. This method has been tested on the Freeman chain code in eight and four directions, the Vertex chain code, the Three OrThogonal chain code, and the Normalized angle difference chain code. The experiments confirmed better compression ratios on various benchmark datasets in comparison to the state-of-the-art lossless chain code compression methods.

Fast Approximate k-Nearest Neighbours Search Using GPGPU

The k-nearest neighbours (k-NN) search is one of the most critical non-parametric methods used in data retrieval and similarity tasks. Over recent years, fast k-NN processing for large amount of high-dimensional data is increasingly demanded. Locality-sensitive hashing is a viable solution for computing fast approximate nearest neighbours (ANN) with reasonable accuracy. This chapter presents a novel parallelisation of the locality-sensitive hashing method using GPGPU, where the multi-probe variant is considered. The method was implemented using CUDA platform for constructing a k-ANN graph. It was compared to the state-of-the-art CPU-based k-ANN and two GPU-based k-NN methods on large and multidimensional data set. The experimental results showed that the proposed method has a speed-up of 30× or higher, in comparison to the CPU-based approximate method, whilst retaining a high recall rate.

GPU-Based Online Optimization of Low Voltage Distribution Network Operation

This paper proposes a parallelized online optimization of low voltage distribution network (LVDN) operation. It is performed on a graphics processing unit (GPU) by combining the optimization procedure with the load flow method. In the case study, performed for the test LVDN with distributed generators (DGs) and controllable loads, differential evolution optimization based on a backward–forward sweep load flow method was parallelized on GPU. The goal of online optimization is to keep the LVDN voltage profile within the prescribed limits, to minimize LVDN losses, and to enable demand response functionality. This is achieved by the optimization determined reference values for the controllable load’s operation, and the reactive power generation, and active power curtailment of DGs. The results show that the parallelized GPU implemented optimization can be significantly faster than similar implementation on a central processing unit, and is, therefore, suitable for the online optimization of the presented LVDN.

Unsigned Manhattan chain code

A new chain code named Unsigned Manhattan Chain Code (UMCC) is presented.UMCC is insensitive to rotation and mirroring, and enables shape magnification.UMCC is capable of detecting monotone parts of the shapes boundary.The proposed chain code achieves 50% better compression ratio than F8 code. This paper introduces a new chain code named Unsigned Manhattan Chain Code - UMCC. Although it exploits a pixels neighbourhood of 8-connectivity, only two coding symbols are used for navigating through the geometric shapes boundary pixels. For this, the movements in the x - and y -coordinate direction are separated, whilst the sign of the moving direction is controlled by two flags - one for each coordinate direction. UMCC is insensitive to rotation and mirroring and enables shape magnification. However, the more unique property is the UMCCs ability to explicitly separate the monotonic parts of the geometric shape. UMCCs properties have been compared against the properties of other chain codes including the Freeman chain code in eight and four directions, the Vertex Chain Code, and the Three OrThogonal chain code. It has been shown that the UMCC has superior properties in regards to the up-to-date chain codes.

Efficient chain code compression with interpolative coding

Abstract This paper considers the use of interpolative coding for lossless chain code compression. The most popular chain codes are used, including Freeman chain code in eight (F8) and four directions (F4), Vertex Chain Code (VCC), and three-orthogonal chain code (3OT). The whole compression pipeline consists of the Burrows–Wheeler transform, Move-To-Front transform and the interpolative coding, which was improved by FELICS and new Ψ-coding. The approach was compared with the state-of-the-art chain code compression algorithms. For VCC, 3OT and F4, the obtained results are slightly better than the existing approaches. However, an important improvement was achieved with F8 chain code, where the presented approach is considerably better.

Visualization and Analytics Tool for Multi-dimensional Data

This paper proposes a novel visualization and analytics tool, which is capable of searching for hidden relationships and patterns within large multi-dimensional data. The goal of the presented tool is to represent the data in novel ways, understandable and useful to the data owner, with new visual and statistical analytics. Various statistics are offered to the user in order to search for linear and nonlinear correlations between multiple variables. Using a simple dataset, we confirmed the suitability of the proposed tool for revealing new relationships and patterns in the used multi-dimensional data.

HELIOS full-waveform laser scanning simulation framework. Source code, precompiled version, example files for study of understory tree height scanning and respective output.

This data collection enables any user to reproduce the study Hammerle et al. (2017). It provides the source code to compile the applied simulation framework. Furthermore, a precompiled version of the software including the necessary files are provided so that a direct start of the simulation without compilation is possible. The output of the simulation (ASCII point clouds) is also provided. For details see the provided README.txt file. The main aim of the study was to examine different laser scanning campaing setups with respect to understory tree height representation which is important, for example, for forestry or climate research.

Photovoltaic potential assessment and ranking of rooftops segments based on LiDAR data

This paper deals with a method for determining the rating of roofs’ segments in urban areas regarding their suitability for the installation of different photovoltaic (PV) systems. In order to determine the received irradiance of individual roofs, their geometry is described based on LiDAR (Light Detection And Ranging) data, in order to estimate accurately the effect of shadowing and topography. The input irradiance is based on a Typical Meteorological Year (TMY), which is established over long-term irradiance measurements. The PV potential is then estimated by integrating estimated per-surface direct and anisotropic diffuse irradiances filtered with nonlinear efficiency characteristics of a given PV system. Afterwards, the roofs’ segments are rated based on the estimated PV potential from low to high suitability. The proposed method was applied over a large urban area scanned by airborne LiDAR, and validated with local power plant, where 96.49% agreement was reached.

Boolean operations on rasterized shapes represented by chain codes using space filling curves

Abstract This paper introduces a new algorithm for Boolean operations on rasterized geometric shapes that are represented with chain codes. The algorithm works in three steps. Firstly, the chain code symbols are transformed in the Hilbert space, where the overlaid chain code symbols are recognised. After that, a suitable starting cell is determined. Finally, the walk-about through the sequence of the initial chain code symbols is performed to obtain the sequence of chain code symbols representing the shape of the required Boolean operation. The algorithm is demonstrated on Freeman chain code in four directions. The time and space complexity of the proposed algorithm is linear, which was proven theoretically and confirmed by experiments.