Thinning and stroke segmentation for handwritten Chinese character recognition

Firstly, a thinning technique by means of stroke tracking is proposed. The method is considered to preserve the straightness of strokes and the length, which is useful for the stroke segmentation procedure on the recognition of handwritten Chinese characters.Secondly, a method for stroke segmentation, i.c. a way of breaking down a character to a set of consecutive partial strokes, is proposed, which works well owing to the favourable properties of the proposed thinning technique. The method consists of five procedures: extraction of feature pixels, calculation of stroke directions, piecewise linear representation of strokes, unification of intersections and extraction of the consecutive partial strokes.     

A cellular logic array for image processing

A cellular logic image processor employing 192 cells in a 16 by 12 hexagonal array is described. The processor has been constructed and its performance assessed. The various classes of functions which can be implemented in the cellular array are discussed and sample programs explained in detail.     

GIAPP: Geological image-analysis program package for estimating geometrical probabilities

When a geological map pattern identifies conditions favorable to mineralization, the geometric probabilities associated with the pattern can be combined with probabilities associated with the distribution of mineral deposits. A tool for estimating these geometric probabilities is provided by GIAPP which can be used as an aid to mineral appraisal.GIAPP uses methods of image processing for computing and displaying interactively the results of transformations of geological binary (0–1) images obtained either by optical or mechanical scanning, or by manual digitization. The data-management system developed for image processing also provides headings with name, type, and dimensions, and comments on the history of processing of each image. Through a simple conversational language, logical operations between images and neighborhood transformations are computed by routines developed for one-dimensional arrays of 1024 bits corresponding to individual rows of image data in binary compressed form. Provision is made for either square or hexagonal raster configurations.Geological maps or textures from thin or polished sections of rocks are displayed on a Tektronix 4014 terminal They are studied interactively by using criteria developed in pattern recognition and in mathematical morphology for texture analysis.     

A technique for encoding lines and regions in engineering drawings

An improvement to a method of extracting line and region information from technical drawings is presented. A scheme for compressing this information as it is being found is also given. The encoding scheme is based on a combination of run-length codes, direction chain codes, and Huffman codes. It is efficient, requires minimal storage, and achieves favorable compression ratios. Experimental results from four typical line drawings are presented.     

A syntax-directed program that performs a three-dimensional perceptual task

A syntax-directed program that performs a three-dimensional perceptual task is described. The task, in a slightly simpler form, was used originally in a psychological study of mental rotation. ⁽¹⁾ The task consists of determining whether two line drawings portray (different views of) identical objects, mirror image objects, or structurally different objects, where the objects are composed of linear strings of attached cubes. The program is syntax-directed in the sense that it uses a fixed set of syntactic rules to analyze the line drawings. This is the first use of formal syntactic techniques in the analysis of pictures (in this case, line drawings) of three-dimensional objects.     

C-calculus: An elementary approach to some problems in pattern recognition

This work utilizes the concept of “Composite set” (C-set) and of the related C-calculus to study some standard problems of pattern analysis and general processing of signals. After some basic definitions and notations about composite sets are briefly stipulated, it is shown how a family of C-sets can be associated with a digitized picture. Each element in the family conveys partial information about the picture itself, yet it is possible to combine the various contributions from each C-set in such a way as to completely retrieve the image. Conditions that guarantee such “convergence” are theoretically investigated: the cases of nonconvergence are also proved to be of some interest.

C-calculus is concretely applied to the extraction of significant regions in a digitized picture, of contours, etc. An application to texture discrimination and analysis is also outlined.     

Extraction of embedded and/or line-touching character-like objects

This paper presents a novel method of circle scanning to detect intersection regions and to extract embedded/line-touching character objects. After the unembedded/not line-touching character objects are trivially extracted, the big line structure is broken down into character-like and line-like segments. Connected component analysis is applied and character-like segments are separated from line-like segments. The character-like segments then pick up the belonging intersection patches to produce character-like objects, leaving the other intersection patches to the line-like segments. Thus the problem of extracting embedded/line-touching character-like objects are addressed.     

Pattern recognition with moment invariants: A comparative study and new results

Moment invariants have been proposed as pattern sensitive features in classification and recognition applications. In this paper, the authors present a comprehensive study of the effectiveness of different moment invariants in pattern recognition applications by considering two sets of data: handwritten numerals and aircrafts.

The authors also present a detailed study of Zernike and pseudo Zernike moment invariants including a new procedure for deriving the moment invariants. In addition, the authors introduce a new normalization scheme that reduces the large dynamic range of these invariants as well as implicit redundancies in these invariants.

Based on a comprehensive study with both handwritten numerals and aircraft data, the authors show that the new method of deriving Zernike moment invariants along with the new normalization scheme yield the best overall performance even when the data are degraded by additive noise.     

On-line recognition (OLREC): A novel approach to visual pattern recognition

A pattern recognition system has been developed which is capable of recognizing high contrast two-dimensional visual patterns and which is insensitive to the translation, rotation and size of a pattern. The unique characteristics of the system are its speed, its low cost and the fact that it is completely self-contained.

A new technique for image recognition, suitable for applications such as industrial robotics, is presented. The technique provides invariance to rotation, translation and magnification of the image. An intelligent camera system is used to input an image, to center it, to normalize it with respect to size and to convert it into polar coordinate form. The image is then integrated over r and θ to provide two orthogonal profiles. The input profiles are compared to the reference images using a threshold comparison technique designed to provide a simple hardware implementation. Rotation of the image is handled by a one-dimensional shifting of the radial profile.

Seven experiments were performed to test the system. The results of these experiments are discussed and possible improvements to the system are suggested.     

Automatic classification of micro-organisms by shape and color, in water quality control: A preliminary study

Automatic segmentation and classification of stained micro-organism images in micrographs are investigated. To discriminate such ellipse-like shapes, a moment based method shows higher reliability than the perimeter based method. A ‘multi-threshold-expansion algorithm’ is successfully applied to segmentation of areas which have darkly stained cores and lightly stained surroundings.     

A highly accurate and computationally efficient approach for unconstrained iris segmentation

Biometric research has experienced significant advances in recent years given the need for more stringent security requirements. More important is the need to overcome the rigid constraints necessitated by the practical implementation of sensible but effective security methods such as iris recognition. An inventive iris acquisition method with less constrained image taking conditions can impose minimal to no constraints on the iris verification and identification process as well as on the subject. Consequently, to provide acceptable measures of accuracy, it is critical for such an iris recognition system to be complemented by a robust iris segmentation approach to overcome various noise effects introduced through image capture under different recording environments and scenarios. This research introduces a robust and fast segmentation approach towards less constrained iris recognition using noisy images contained in the UBIRIS.v2 database (the second version of the UBIRIS noisy iris database). The proposed algorithm consists of five steps, which include: (1) detecting the approximate localization of the eye area of the noisy image captured at the visible wavelength using the extracted sclera area, (2) defining the outer iris boundary which is the boundary between iris and sclera, (3) detecting the upper and lower eyelids, (4) conducting the verification and correction for outer iris boundary detection and (5) detecting the pupil area and eyelashes and providing means for verification of the reliability of the segmentation results. The results demonstrate that the accuracy is estimated as 98% when using 500 randomly selected images from the UBIRIS.v2 partial database, and estimated at ?97%$97\%$ in a “Noisy Iris Challenge Evaluation (NICE.I)” in an international competition that involved 97 participants worldwide, ranking this research group in sixth position. This accuracy is achieved with a processing speed nearing real time.     

A novel, rule-based technical pattern identification mechanism: Identifying and evaluating saucers and resistant levels in the US stock market

This paper has two main purposes. The first one is the development of a rigorous rule-based mechanism for identifying the rounding bottoms (also known as saucers) pattern and resistant levels. The design of this model is based solely on principles of technical analysis, and thus making it a proper system for evaluating the efficacy of the aforementioned technical trading patterns. The second aim of this paper is measuring the predictive power of buy-signals generated by these technical patterns. Empirical results obtained from seven US tech stocks indicate that simple resistant levels outperform saucers patterns. Furthermore, positive statistical significant excess returns are being generated only in first sub-periods of examination. These returns decline or even vanish as the experiment proceeds to recent years. Our findings are aligned with the results reported by various former studies. The proposed identification mechanism can be used as a component of an expert system to assist academic community in evaluating trading strategies where technical patterns are embedded.     

Distance and similarity measures between intuitionistic fuzzy sets: A comparative analysis from a pattern recognition point of view

A detailed analysis of the distance and similarity measures for intuitionistic fuzzy sets proposed in the past is presented in this paper. This study aims to highlight the main theoretical and computational properties of the measures under study, while the relationships between them are also investigated. Along with the literature review, a comparison of the analyzed distance and similarity measures from a pattern recognition point of view in three different classification cases is also presented. Initially, some artificial counter-intuitive recognition cases are considered, while in a second phase real data from medical and well known pattern recognition benchmark problems are used to examine the discrimination abilities of the studied measures. Moreover, all the measures are applied in a face recognition problem for the first time and useful conclusions are drawn regarding the accuracy and confidence of the recognition results. Finally, the measures’ suitability and their drawbacks that make the development of more robust and efficient measures’ a still open issue are discussed.     

A two-level committee machine: a representation and a learning procedure for general piecewise linear discriminant functions

In order to improve the pattern classification power of the committee machine, a two-level committee machine, which is a committee machine with several lower committees, is proposed and the learning algorithm for it is described. The discriminant function realized by the two-level committee machine can be considered as the general piecewise linear discriminant function which includes Chang's definition.⁽¹⁵⁾ The proposed algorithm is a kind of error-correction procedure, and the learning procedure of the usual committee machine and the perceptron are clearly explained as special cases of the proposed algorithm.     

MIPAS: A neural network-based multimedia information processing and analysis system

This paper presents the structure and design criteria of a neural network-based multimedia information processing and analysis system (MIPAS) which can be used to deal with more-complicated intelligence issues. According to the structure and design criteria, a software environment (SEMIPAS), which supports the implementation of multimedia information (image + speech, image + characters, speech + characters, image + speech + characters) processing and analysis applications, is implemented and introduced. Under this software environment, a multimedia information processing and analysis system called “To Know the World” is constructed. Experiments show that the multimedia information processing and analysis is much more powerful and effective than single-medium information processing and analysis.     

A new feature extractor invariant to intensity, rotation, and scaling of color images

This paper proposes a new method for extracting the invariant features of an image based on the concept of principal component analysis and a competitive learning algorithm. The proposed algorithm can be applied to binary, gray-level, or colored-texture images with a size greater than 256 × 256 pixels. In addition to translation, scaling, and rotation invariant extraction, the extraction of a feature invariant to color intensity can be implemented by using this method. In our experiment, the proposed method shows the capability to differentiate images having the same shape but different colored textures. The experimental results report the effectiveness of this technique and its performance as measured by recognition accuracy rate and computational time. These results are also compared with those obtained by classical techniques.     

MRCCA: A novel CCA based method and its application in feature extraction and fusion for matrix data

Multiset features extracted from the same pattern usually represent different characteristics of data, meanwhile, matrices or 2-order tensors are common forms of data in real applications. Hence, how to extract multiset features from matrix data is an important research topic for pattern recognition. In this paper, by analyzing the relationship between CCA and 2D-CCA, a novel feature extraction method called multiple rank canonical correlation analysis (MRCCA) is proposed, which is an extension of 2D-CCA. Different from CCA and 2D-CCA, in MRCCA k pairs left transforms and k pairs right transforms are sought to maximize correlation. Besides, the multiset version of MRCCA termed as multiple rank multiset canonical correlation analysis (MRMCCA) is also developed. Experimental results on five real-world data sets demonstrate the viability of the formulation, they also show that the recognition rate of our method is higher than other methods and the computing time is competitive.