Parallel region growing of half-tone images based on selected average brightness of the area along the growth route

The problem of parallel segmentation of halftone images by brightness for implementation on the basis of programmable logic integrated circuits is considered. Segmentation divides an image into regions formed from pixels with approximately the same brightness, and is a computationally complex operation due to multiple checks of the value of each pixel for the possibility of joining an adjacent region. To speed up segmentation, parallel algorithms for growing areas have been developed, in which processing begins from the neighborhoods of preselected initial growth pixels. The condition of joining an adjacent pixel to an area takes into account the average brightness of the area to limit the variance of its pixel values. Therefore, when each new pixel is added to the area, its average brightness is recalculated. This leads to high time complexity. In some parallel algorithms, the sample mean is calculated in a small window, which makes it possible to slightly reduce the time complexity when matching the window size with the segment sizes. To significantly reduce the temporal complexity, the article proposes a model for the parallel growth of image regions based on a simplified condition for joining adjacent pixels to a region, taking into account the sample average value of the region's brightness along the growth route connecting the boundary pixel of the region and the initial growth pixel through a sequence of pixels used to attach the considered boundary pixel to area. A significant decrease in the temporal complexity of the proposed model of parallel growing of image regions in comparison with the known models is achieved due to a slight increase in the spatial complexity.

A Numerical Method to Improve the Representativeness of Real Microstructure Cut-Outs Applied in Finite Element Simulations

To improve the representativeness of a real microstructural cut-out for modeling purposes, a numerical method named as “boundary pixel color alteration (BPCA)” is presented to modify measured 2D microstructure cut-outs. Its physical background is related to the phase growth. For the application, the precondition is that the representativeness of the microstructure is already satisfied to a certain extent. This method resolves the problem that the phase composition of a small cut-out can have a large discrepancy to the real one. The main idea is to change the pixel color among neighboring pixels belonging to different phases. Our process simultaneously maintains most of the characteristics of the original morphology and is applicable for nearly all kinds of multi-phase or polycrystalline metallic alloys, as well. From our axisymmetric finite element (FE) simulations (ABAQUS ) applied with 2D real microstructures, it shows that the volume ratios of microstructural phases, as a function of the structure position to the symmetric axis, converge to phase area ratios in the 2D cut-out, even though the axisymmetric element volume is position dependent. A mathematical proof provides the reason for the aforementioned convergence. As examples to achieve real compositions and to numerically prove the aforementioned convergence, four different materials including multiphase polycrystals are implemented. An improvement of the predicted FE result is presented for the application of a modified microstructure (with a higher representativeness) compared to the original one.

Off-line signature verification using elementary combinations of directional codes from boundary pixels

Verifying the genuineness of official documents, such as bank checks, certificates, contract forms, bonds, etc., remains a challenging task when it comes to accuracy and robustness. Here, the genuineness is related to the degree of match of the signature contained in the documents relating to the original signatures of the authorized person. Signatures of authorized persons are considered known in advance. In this paper, a novel feature set is introduced based on quasi-straightness of boundary pixel runs for signature verification. We extract the quasi-straight line segments using elementary combinations of the directional codes from the signature boundary pixels and subsequently we obtain the feature set from various quasi-straight line classes. The quasi-straight line segments provide a blending of straightness and small curvatures resulting in a robust feature set for the verification of signatures. We have used Support Vector Machine (SVM) for classification and have shown results on standard signature datasets like CEDAR (Center of Excellence for Document Analysis and Recognition) and GPDS-100 (Grupo de Procesado Digital de la Senal). The results establish how the proposed method outperforms the existing state of the art.

Semantic 3D Mapping from Deep Image Segmentation

The perception and identification of visual stimuli from the environment is a fundamental capacity of autonomous mobile robots. Current deep learning techniques make it possible to identify and segment objects of interest in an image. This paper presents a novel algorithm to segment the object’s space from a deep segmentation of an image taken by a 3D camera. The proposed approach solves the boundary pixel problem that appears when a direct mapping from segmented pixels to their correspondence in the point cloud is used. We validate our approach by comparing baseline approaches using real images taken by a 3D camera, showing that our method outperforms their results in terms of accuracy and reliability. As an application of the proposed algorithm, we present a semantic mapping approach for a mobile robot’s indoor environments.

Graph-based segmentation with homogeneous hue and texture vertices

This work presents an automated segmentation method, based on graph theory, which processes superpixels that exhibit spatially similarities in hue and texture pixel groups, rather than individual pixels. The graph shortest path includes a chain of neighboring superpixels which have minimal intensity changes. This method reduces graphics computational complexity because it provides large decreases in the number of vertices as the superpixel size increases. For the starting vertex prediction, the boundary pixel in first column which is included in this starting vertex is predicted by a trained deep neural network formulated as a regression task. By formulating the problem as a regression scheme, the computational burden is decreased in comparison with classifying each pixel in the entire image. This feasibility approach, when applied as a preliminary study in electron microscopy and optical coherence tomography images, demonstrated high measures of accuracy: 0.9670 for the electron microscopy image and 0.9930 for vitreous/nerve-fiber and inner-segment/outer-segment layer segmentations in the optical coherence tomography image.

Identifikasi Nilai Piksel Sepanjang Garis Pantai di Teluk Genteng Menggunakan Citra Landsat 4-5 TM

ABSTRAKTeknologi pengindraan jauh memudahkan dalam penyediaan data untuk memperoleh garis pantai, terutama untuk wilayah lautan yang luas seperti Indonesia. Ekstraksi garis pantai dari citra satelit merupakan salah satu cara untuk memperoleh nilai-nilai piksel sepanjang garis pantai. Tujuan dari penelitian ini adalah mencari dan menentukan pola sebaran nilai piksel di sepanjang garis pantai area sampel di Teluk Genteng serta untuk mengetahui sinkronisasi dengan garis pantai dari Peta RBI. Penelitian ini menggunakan citra satelit Landsat 4-5 TM dengan Band 5 digunakan untuk memisahkan darat dan air. Pola sebaran nilai piksel didapatkan dengan menampalkan hasil ekstraksi dengan citra asli dan Peta RBI. Dari penelitian ini diperoleh bahwa pola sebaran nilai piksel batas yang terjadi pada area sampel menyebar secara acak, nilai piksel batas tertinggi adalah 242, nilai piksel batas terendah adalah 27, rentang nilai piksel batas adalah 215, rentang nilai piksel air adalah 126, rentang nilai piksel darat adalah 143, dan secara umum bentuk dari ekstraksi hampir mendekati garis pantai Peta RBI.Kata kunci: ekstraksi garis pantai, nilai piksel, Peta RBIABSTRACTTechnologies in remote sensing give the effortless way in preparing data to build the coastline, especially for country which has vast areas of sea such as Indonesia. Coastline extraction from satellite imagery is one of alternatives to get the pixel values along the coastline. To find and determine the distribution pattern of pixel values along the coastline of sampel area in Genteng Bay and to determine the synchronization with the coastline of RBI Map are taken as the focus of this research. Landsat 4-5 TM with Band 5 is the satellite imagery which is used to separate land and water. The distribution pattern of pixel values is obtained through overlaying the extraction with the initial pixel values and RBI Map. It can be detected that distribution of boundary pixel values in sampel area show the random pattern, the maximum values is 242, while the minimum is 27 with range of 215. The range of water pixel values is 126 and 143 for the land. Besides, the coastline extraction almost coincides with RBI Map coastline.Keywords: extraction of coastline, pixel values, RBI Map

An Improved Video Mosaic Block Detection Method on the Statistics of Pixel Difference across the Boundary of Macroblock and Suspected Mosaic Blocks

This passage proposes a new method to detect mosaic not only using the Y (luminance) component in YUV color space of videos, but also using the U (chrominance) and V component. The mosaic effect is measured by the boundary pixel difference from the neighbor macroblock. Instead of detecting the existence and position of mosaic blocks as traditional methods do, this method focuses on the statistics of the number of suspected mosaic blocks so that the quality of the whole frame affected by mosaic can be assessed. Experimental results show that the new method has good performance on fallout ratio, omission factor and computational complexity.

Ambiguous Region Matching and Boundary Pixel Optimization for Stereo Computation

In this paper, we present a stereo matching algorithm based on planar surface hypothesis. It improves the results of low texture regions and mixed pixels on object boundaries. First, regions are segmented by applying the mean-shift segmentation method. Then we propose a coarse-to-fine algorithm to increase the reliable correspondences in low texture regions. Third, the Belief Propagation algorithm is used to optimize disparity plane labeling. Finally, for a mixed pixel, we utilize the results of the depth plane and the local region of it to regulate its disparity. Experimental results using the Middlebury stereo test show that the performance of our method is high.