Image segmentation scheme based on SOM–PCNN in frequency domain

A hybrid scheme for the image segmentation of high-resolution images is proposed in this study. Our methodology is based on combining both supervised and unsupervised segmentation. The entire process is performed in the frequency domain, rather than the spatial domain, using the Shift Invariant Shearlet Transform (SIST). Initially, the input image is filtered using an anisotropic filter to enhance the texture features. Then, it is separated into low and high sub-band frequencies using SIST. Subsequently, we built a feature vector from coarser coefficients complemented with texture information extracted from high-frequency coefficients of the input image. SOM is used for the preliminary classification of the input image coefficients, and the network training process is performed using the previously built feature vector. Lastly, the modified PCNN is used to augment the SOM results to reduce the over-segmentation artefacts. We used the Berkeley Segmentation Database (BSR) and Quick-Bird Satellite images to validate the results. It was found that the proposed scheme is superior to the Fuzzy-C-Means-based, SOM-based, and PCNN-based segmentation algorithms in terms of quantitative criteria and visual interpretation.     

Ultrasonic Tomographic Imaging of an Encased,Highly Attenuating Solid Medium

Abstract

Ultrasonic tomographic imaging has been performed on a highly attenuating, filled, hydroxyl-terminated polybutadiene, cast into a steel case. The detection of defects and the characterization of solid rocket propellant were thus possible. Extremely low signal levels due to the attenuation and scattering properties of the polymer were present. In addition, a nonideal tomographic reconstruction geometry was present, because of a cylindrical air-filled cavity in the center of the polymer. Complications also arose from high-amplitude guided waves within the steel case, which tended to mask directly transmitted signals through the propellant. These problems were solved using specially designed transducers and pulse-compression signal processing techniques. Modifications to the filtered back-projection reconstruction procedure were made to account for the discontinuity caused by the hole. Images are presented showing the detection of an artificial defect to demonstrate the technique.     

Multiscale texture element detection

A multiscale algorithm is described, mathematically justified, and demonstrated, which enables extraction of various sizes and shapes of texture elements from natural images without need for tuning parameters.     

Hierarchical image segmentation based on similarity of NDVI time series

Although a variety of hierarchical image segmentation procedures for remote sensing imagery have been published, none of them specifically integrates remote sensing time series in spatial or hierarchical segmentation concepts. However, this integration is important for the analysis of ecosystems which are hierarchical in nature, with different ecological processes occurring at different spatial and temporal scales. Therefore, the objective of this paper is to introduce a multi-temporal hierarchical image segmentation (MTHIS) methodology to generate a hierarchical set of segments based on spatial similarity of remote sensing time series. MTHIS employs the similarity of the fast Fourier transform (FFT) components of multi-seasonal time series to group pixels with similar temporal behavior into hierarchical segments at different scales. Use of the FFT allows the distinction between noise and vegetation related signals and increases the computational efficiency. The MTHIS methodology is demonstrated on the area of South Africa in an MTHIS protocol for Normalized Difference Vegetation Index (NDVI) time series. Firstly, the FFT components that express the major spatio-temporal variation in the NDVI time series, the average and annual term, are selected and the segmentation is performed based on these components. Secondly, the results are visualized by means of a boundary stability image that confirms the accuracy of the algorithm to spatially group pixels at different scale levels. Finally, the segmentation optimum is determined based on discrepancy measures which illustrate the correspondence of the applied MTHIS output with landcover-landuse maps describing the actual vegetation. In future research, MTHIS can be used to analyze the spatial and hierarchical structure of any type of remote sensing time series and their relation to ecosystem processes.     

Bin-picking with a solid state range camera

In this paper we present new work done on the bin-picking problem. The work was triggered by the advent of a new solid state range camera which enables the economic and robust use of range imagery in industrial robotic automation tasks. The application presented is that of pick-and-place of randomly oriented but known polyhedral objects in an industrial robotic work cell. The algorithms for segmentation, pose estimation, and grasp point determination are presented along with practical results from a real industrial grade work cell.     

A fast iterative scheme for multilevel thresholding methods

The previously published optimal thresholding techniques based on some objective functions are very efficient in the bi-level thresholding case, but they are impractical when extended to multilevel thresholding. The reason for this is their computational complexity which grows exponentially with the number of thresholds. In this paper, an iterative scheme is proposed to render these optimal thresholding techniques more practical. The proposed algorithm starts with a bi-level thresholding, then uses the initial results to obtain higher-order thresholds. This algorithm is iterative and the convergence is proved. We also introduce some useful programming techniques to make the computation more efficient. The proposed algorithm can therefore determine the number of thresholds automatically as well as save a significant amount of computing time.     

On the interaction of channeling/segmentation cracks of coating: Study on the critical spacing

Channeling/segmentation cracks may arise in the coating subjected to in-plane tensile stress. The interaction between these multiple cracks, say the effect of the spacing between two adjacent cracks on the behaviors of channels themselves and the interface around the interface corners, attracts wide interest. However, if the spacing is greater than a specific magnitude, namely the Critical Spacing (CS), there should be no interaction between such channeling/segmentation cracks. In this study, the mechanism of the effect of the crack spacing on the interfacial stress around the interface corner will be interpreted firstly. Then the existence of the CS will be verified and the relationship between the CS and the so-called stress transfer length in coating will be established for plane strain condition. Finally the dependence of the stress transfer length, simultaneously of the CS, on the sensitive parameters will be investigated with finite element method and expressed with a simple empirical formula.     

基于分层遗传算法的双阈值图像分割

图像分割是图像处理中的重要问题,通常的图像分割法包括阈值法、边缘检测法、区域跟踪法等.其中,阈值法是一种通用的方法.本文将分层遗传算法用于阈值的选取,仿真结果表明,在设定了合适的遗传算子后,该算法可以实现正确的图像分割,将分层遗传算法用于图像处理中,是非常有效的.     

VLSI design and implementation of a real-time image segmentation processor

Image segmentation is a crucial part of machine vision applications. In this paper a system to perform real-time segmentation of images is presented. It uses a real-time segmentation VLSI chip that is based on a gradient relaxation algorithm and is designed using the Path Programmable Logic design methodology developed at the University of Utah. The system design considerations, system specifications, and an input/output format for the chip are discussed. The actual design of the chip is given that uses pipeline methodology to achieve real-time performance with a compact VLSI layout. The implementation of the segmentation system is presented and the segmentation chip and the overall system are evaluated with regard to real-time performance and segmentation results.This work was supported in part by Grant ISI-856-0393 from the National Science Foundation.