John Cowell

The Java Language

The Java programming language is similar to the C++ language - but with some interesting differences. If you are a C++ expert you will be able to zip through this chapter quite fast - chapter 15 deals specifically with the differences between the two languages. If you have used a high level language such as Pascal before you will also find the switch to Java straightforward.

A fast recognition system for isolated arabic characters

This paper presents a very fast multi-stage algorithm for the recognition of non-Latin script. Although the examples use Arabic script, the system could be adapted in minutes to deal with any character set, in particular non-Latin characters where no commercial OCR systems are available. The approach used normalises isolated characters for size and extracts an image signature based on the number of black pixels in the rows and columns of the character and compares these values to a set of signatures for typical characters of the set. This technique identifies not only the closet match but gives the closeness of match to all other characters in the set, which is expressed in a triangular confusion matrix.

Extracting subtle facial expression for emotional analysis

Humans convey emotions with subtle facial expressions. The interpretation of these expressions not only depends on that expression, but on how it developed. Exposing this development enables a better analysis of the ambiguity experienced when analyzing the underlying emotion. Subtle facial expressions can be objectively measured in terms of action units as defined in the facial action coding system. This paper presents a facial expression language based on FACS which provides a continues stream of expressions for measuring the growth of expression. The development of the emotion surprise which requires two action units is used as a test case

Thinning Arabic characters for feature extraction

A successful approach to the recognition of Latin characters is to extract features from that character such as the number of strokes, stroke intersections and holes, and to use ad-hoc tests to differentiate between characters which have similar features. The first stage in this process is to produce thinned 1 pixel thick representations of the characters to simplify feature extraction. This approach works well with printed Latin characters which are of high quality. With poor quality characters, however, the thinning process itself is not straightforward and can introduce errors which are manifested in the later stages of the recognition process. The recognition of poor quality Arabic characters is a particular problem since the characters are calligraphic with printed characters having widely varying stroke thicknesses to simulate the drawing of the character with a calligraphy pen or brush. This paper describes the problems encountered when thinning large poor quality Arabic characters prior to the extraction of their features and submission to a syntactic recognition system.

Character recognition of Arabic and Latin scripts

The goal to produce effective optical character recognition (OCR) methods has led to the development of a number of algorithms. The purpose of these is to take the handwritten or printed text and to translate it into a corresponding digital form. The multitude requirements and developments are well represented in the literature (I.S.I. Abuhaiba et al., 1994: C.Y. Suen, 1986). The primary objective of the paper is to provide an insight into a robust system which has been successfully developed and employed to recognise Latin and Arabic characters and whose workings has been described previoulsy (J. Cowell and F. Hussain, 2000). The focus is to discuss the main components used in the multi-stage system, paying particular attention to the normalisation process used for orientation and size for a given bitmapped character. The effectiveness of the approach is demonstrated through its workings for the Arabic and Latin case, both for characters and numbers.

Amharic character recognition using a fast signature based algorithm

The Amharic language is the principal language of over 20 million people mainly in Ethiopia. An extensive literature survey reveals no journal or conference papers on Amharic character recognition. The Amharic script has 33 basic characters each with seven orders giving 231 distinct characters, not including numbers and punctuation symbols. The characters are cursive but not connected and unlike other cursive scripts do not use dots. We describe the Amharic script and discuss the difficulties of applying conventional structural and syntactic recognition processes. Two statistical algorithms for identifying Amharic characters are described. In both, the characters are normalised for both size and orientation. The first compares the character against a series of templates. The second derives a characteristic signature from the character and compares this against a set of signature templates. The signatures used are fifty times smaller than the original character and the recognition process is corresponding faster but with some loss of accuracy. The statistical techniques described have been fully implemented and the resulting performance outlined.

Emotional analysis of facial expressions

The purpose of this work is to develop an emotional analyzer that enables a cognitive agent to determine the emotional state of another cognitive agent through facial expressions. These expressions are presented to the analyzer using a facial expressions language. The analyzer fuses these formulas to determine the basic emotions associated with these formulas and eventually derive the emotional state. The main characteristic of this analyzer is that it takes the current emotional state and internal state into consideration during the emotional state calculation. In this paper, we present the concepts and algorithms used to develop facial expressions emotions analyzer (FEEA)

Java and C

When a C++ programmer first see Java, the immediate reaction is that Java is really the same as C++. While the two languages have much in common there are some significant differences between them that can cause problems if you are not aware of them.

A need for improved standardization of metadata for consumer devices

There is an urgent need for metadata to accompany and describe media data essence. With the growing amount of personal content, user generated descriptors tend to be vague to the point of uselessness. There is a need for standardization in the metadata format to allow a full and useful description of content that is interoperable between consumer devices. Manufacturers need to ensure that the metadata generated by a device is complete and understood by other products as well as allow for the creation and use of more subjective metadata. This paper provides a background on metadata levels and existing metadata standards with the aim to show deficiencies within current standards and to indicate the need for improved standardization.

Functions and Procedures

In a Windows environment whenever an event occurs a block of Visual Basic statements called an event handler is executed. An event handler is a procedure bound to an event. We have seen this in every application we have developed so far. In addition to using these procedures we can create our own. A procedure is just a collection of Visual Basic statements, in the case of event procedures they are linked to visual controls, while any other procedures we create are not. The advantage of creating new procedures rather than just writing all the Visual Basic code in event procedures is that a procedure may be called from more than one place, which reduces the need to duplicate code. Wherever possible it is best to reuse code as much as possible since it makes applications less complicated to test and reduces the number of possible errors. A function is very similar to a procedure except that it returns a value. In this chapter we are going to see how to create and use procedures and functions.