Wojciech Bieniecki

Image Preprocessing for Improving OCR Accuracy

Digital cameras are convenient image acquisition devices: they are fast, versatile, mobile, do not touch the object, and are relatively cheap. In OCR applications, however, digital cameras suffer from a number of limitations, like geometrical distortions. In this paper, we deal with the preprocessing step before text recognition, specifically with images from a digital camera. Experiments, performed with the FineReader 7.0 software as the back-end recognition tool, confirm importance of image preprocessing in OCR applications.

Pretransplantation Cellular Alloreactivity Is Predictive of Acute Graft Rejection and 1-Year Graft Function in Kidney Transplant Recipients

OBJECTIVE To study cellular alloimmunity in kidney allograft recipients using an interferon-gamma enzyme-linked immunosorbent spot assay (ELISPOT). MATERIAL AND METHODS Donor splenocyte peripheral blood mononuclear cells were obtained during kidney recovery in 53 kidney recipients including 11 with positive panel-reactive antibodies pretransplantation. For ELISPOT data analysis, the spot number, size, and intensity were calculated, reflecting the volume of cytokine secretion at the single-cell level. Results were recalculated as the ratio of the values observed for donor-stimulated to unstimulated recipient cells corrected for residual donor activity. RESULTS Significantly greater pretransplantation donor-stimulated activity was observed in recipients who experienced an acute rejection episode (ARE) within 1 year (P < .05). Mean change in spot number, size, and intensity in patients without or with AREs was 0.99 vs 3.33, 1.60 vs 6.05, and 1.40 vs 6.31, respectively. The assessed parameters were prognostic of high risk of ARE: 1.5-fold increase in spot number (ARE incidence, 52% vs 9%), 2.5-fold increase in spot size (ARE incidence, 53% vs 13%), and 2.7-fold increase in spot intensity (ARE incidence, 52% vs 9%). The 3 parameters correlated with 1-year serum creatinine concentration (P < .05). In 14 recipients, AREs could have been predicted in 11 using pretransplantation ELISPOT results, and in only 2 on the basis of panel-reactive antibodies. CONCLUSION The ELISPOT-determined capacity of donor-induced reactivity observed in recipient cells obtained just before transplantation is predictive of risk of graft rejection and 1-year allograft function.

Merging Adjacency Lists for Efficient Web Graph Compression

Analysing Web graphs meets a difficulty in the necessity of storing a major part of huge graphs in the external memory, which prevents efficient random access to edge (hyperlink) lists. A number of algorithms involving compression techniques have thus been presented, to represent Web graphs succinctly but also providing random access. Our algorithm belongs to this category. It works on contiguous blocks of adjacency lists, and its key mechanism is merging the block into a single ordered list. This method achieves compression ratios much better than most methods known from the literature at rather competitive access times.

Tight and simple Web graph compression for forward and reverse neighbor queries

Analyzing Web graphs has applications in determining page ranks, fighting Web spam, detecting communities and mirror sites, and more. This study is however hampered by the necessity of storing a major part of huge graphs in the external memory which prevents efficient random access to edge (hyperlink) lists. A number of algorithms involving compression techniques have thus been presented, to represent Web graphs succinctly, but also providing random access. Those techniques are usually based on differential encodings of the adjacency lists, finding repeating nodes or node regions in the successive lists, more general grammar-based transformations or 2-dimensional representations of the binary matrix of the graph. In this paper we present three Web graph compression algorithms. The first can be seen as engineering of the Boldi and Vigna (2004) [8] method. We extend the notion of similarity between link lists and use a more compact encoding of residuals. The algorithm works on blocks of varying size (in the number of input lists) and sacrifices access time for better compression ratio, achieving more succinct graph representation than other algorithms reported in the literature. The second algorithm works on blocks of the same size in the number of input lists. Its key mechanism is merging the block into a single ordered list. This method achieves much more attractive space-time tradeoffs. Finally, we present an algorithm for bidirectional neighbor query support, which offers compression ratios better than those known from the literature.

A two-pass median filter for impulse noise attenuation in color images

We propose a two-pass median filter for impulse noise suppression in color images. The first pass is our previously presented algorithm, PNN-VMF. The second pass routine is chosen dynamically on the basis of the number of pixel modifications performed by the first pass filtering. Such an approach is likely to avoid barm (blurring edges etc.) due to excessive filtering. The effectiveness of the algorithm has been experimentally verified.

Automatic segmentation and recognition of pathomorphological microscopic images

This paper is concerned with quantitative analysis of color biomedical images. We take into account microscopic cytological and histological images. The aim of the examination is to extract two types of object from a set of images and evaluate their count and area. An essential step in the recognition procedure is image segmentation, due to the poor quality of the source images. We test some segmentation techniques including the combination of color discrimination and region based methods.

The Study of Requirements for the System of Automatic Measurement of Vegetation Cover in River Catchments

In this paper we present an application of statistical pattern recognition methods to the analysis of air photographs in order to evaluate the vegetation cover in a Pilica River catchment (Central Poland). The analysis is a basis for using plants to improving water quality and increasing its availability. The task for the image processing and analysis system is to recognize in the air picture the regions of different vegetation forms and water containers, and evaluate their area. The most difficult problem from the point of image processing is segmentation due to altering the color of vegetation forms. An automatic image analysis is intended to replace geodetic measurements.

Preprocessing for Real-Time Handwritten Character Recognition

We present a real-time on-line handwritten character recognition system, based on an ensemble of neural networks. In this work we focus on the developed preprocessing algorithms which help achieve high accuracy rate without a visible delay in recognition process.

copMEM: finding maximal exact matches via sampling both genomes

Motivation: Genome‐to‐genome comparisons require designating anchor points, which are given by Maximum Exact Matches (MEMs) between their sequences. For large genomes this is a challenging problem and the performance of existing solutions, even in parallel regimes, is not quite satisfactory. Results: We present a new algorithm, copMEM, that allows to sparsely sample both input genomes, with sampling steps being coprime. Despite being a single‐threaded implementation, copMEM computes all MEMs of minimum length 100 between the human and mouse genomes in less than 2 minutes, using 7 GB of RAM memory. Availability and implementation: https://github.com/wbieniec/copmem Supplementary data: Supplementary data are available at Bioinformatics online.

Identification and Assessment of Selected Handwritten Function Graphs Using Least Square Approximation Combined with General Hough Transform

Abstract The paper provides a comparison of three variants of algorithms for automatic assessment of some examination tasks involving sketching a function graph based on image processing. Three types of functions have been considered: linear, quadratic, and trigonometric. The assumption adopted in the design of the algorithm is to map the way the examiner assesses the solutions and to achieve the evaluation quality close to the one obtained in manual evaluation. In particular, the algorithm should not reject a partly correct solution and also extract the correct solution from other lines, deletions and corrections made by a student. Essential subproblems to solve in our scheme concern image segmentation, object identification and automatic understanding. We consider several techniques based on Hough Transform, least square fitting and nearest neighbor based classification. The most reliable solution is an algorithm combining least square fitting and Hough Transform.