Narrow Results

The contours and segments of objects in digital images have many important applications. Contour extractions of gray images can be converted into contour extractions of binary images. This paper presents a novel contour-extraction algorithm for binary images and provides a deduction theory for this algorithm. First, we discuss the method used to construct convex hulls of regions of objects. The contour of an object evolves from a convex polygon until the exact boundary is obtained. Second, the projection methods from lines to objects are studied, in which, a polygon iteration method is presented using linear projection. The result of the iteration is the contour of the object region. Lastly, addressing the problem that direct projections probably cannot find correct projection points, an effective discrete ray-projection method is presented. Comparisons with other contour deformation algorithms show that the algorithm in the present paper is very robust with respect to the shapes of the object regions. Numerical tests show that time consumption is primarily concentrated on convex hull computation, and the implementation efficiency of the program can satisfy the requirement of interactive operations.

Studying topological characteristics of digital images is a fundamental issue in image analysis and understanding. In the present paper we first propose a linear time constant-working space algorithm for determining the genus of a connected digital image. The computation is based on a combinatorial relation for digital images that may be of independent interest as well. We also propose definitions of dimension for planar digital images. These definitions serve as an alternative to the one proposed by Mylopoulos and Pavlidis¹, and make up some of its shortcomings. We study various properties of the so-defined image dimension, in particular, characterization of dimension in terms of Euler characteristic. We also show that image dimension can be found within linear time and constant memory.

Security of digital data is an important task in the present era. In this paper, we propose a new scheme of digital image encryption and decryption method based on three-dimensional (3D) Arnold cat map (ACM) and elliptic curve. In this proposed encryption method, we have applied 3D ACM on the digital color image which performs the dual encryption first, it performs the permutation and second, it performs the substitution of image pixels. After that, elliptic curve cryptography (ECC) is used to encrypt the image, for this a mapping method is proposed to convert the pixels of the image as points on the elliptic curve. Further, a mapping inverting method is proposed for decryption and then 3D inverse Arnold cat map (iACM) is applied to get the original image. The statistical and security analyses are done on various images and the experimental results show the robustness of the proposed method.

Perhaps the best known example of user-guided evolution is furnished by evolving expressions, an image generation technique first introduced by Sims. In this version of artificial evolution, images are evolved for aesthetic purposes, hence any fitness measure used must be based on aesthetics. We consider the problem of guiding image evolution autonomously on the basis of computational, as opposed to user-assigned, aesthetic fitness. Due to the difficulty of formulating an absolute criterion for aesthetic fitness, we adopt a coevolutionary approach, relying on hosts and parasites to establish relative criteria for aesthetic fitness. To sustain the coevolutionary arms race, we allow coevolution to proceed in stages. This permits appropriate fitness levels to be maintained within the parasite populations we use to infect host image populations. Using staged coevolution produces two beneficial results: (1) longer survival times for subpopulations of host images, and (2) stable phenotypic lineages for host images.

The numerical analysis technique came to be used for elucidation to various problems of the engineering field thanks to rapid advancement of the computing technology and popularization of computer. To simulate today's problems accurately, some large-scale analysis must be carried out. However, the problem of the input data making is pointed out in Finite Element Method (FEM) etc. Such a situation is received, meshless method is researched actively in a variety of engineering fields. There is Free Mesh Method (FMM) in a kind of these techniques. FMM does not require any connectivity between nodes and elements for an input data and is able to obtain accuracy which is almost equal to FEM. Authors applied the tensile fracture analysis of plane concrete using FMM in 2 dimension, and obtained excellent results. However, actual concrete structures have 3 dimensional behaviors, and the fracture behavior is extremely complex. In addition, the fracture behaviors of concrete are not analyzed enough in 3 dimension. This paper describes an application of FMM in 3 dimension to complex fracture of concrete, and some examples of the numerical analysis are shown and excellent results are obtained.