A computer aided local estimation of fusion splice loss for graded-index optical fibers is proposed. Spliced fiber outer diameter deformation is extracted by image processing technique. The relation between deformation at the splice portion and splice loss is obtained experimentally. It is possible to eliminate large loss splices automatically using the splice portion appearance.

An integrated pen interface system was developed to allow effective Japanese text entry. It consists of sub-systems for handwriting recognition, contextual post-processing, and enhanced Kana-to-Kanji conversion. The recognition sub-system uses a hybrid algorithm consisting of a pattern matcher and a neural network discriminator. Special care was taken to improve the recognition of non-Kanji and simple Kanji characters frequently used in fast data entry. The post-processor predicts consecutive characters on the basis of bigrams modified by the addition of parts of speech and substitution of macro characters for Kanji characters. A Kana-to Kanji conversion method designed for ease of use with a pen interface has also been integrated into the system. In an experiment in which 2,900 samples of Kanji and non-Kanji characters were obtained from 20 subjects, it was observed that the original recognition accuracy of 83.7% (the result obtained by using the pattern matching recognizer) was improved to 90.7% by adding the neural network discriminator, and that it was further improved to 94.4% by adding the post-processor. The improved recognition accuracy for non-Kanji characters was particularly marked.

Fractals have been widely applied to picture processings and related fields. They have been particularly successful as computer graphics tools. However, fractal applications to image analysis have not been successful. Because,conventional fractal feature--fractal dimension (F-dim)--is scalar valued, and it is difficult to characterize various image information using the conventional F-dim. In this paper, a new concept of a fractal feature is proposed to overcome the above drawback. The conventional F-dim satisfies a kind of scaling equation. The new fractal feature is a generalization of the conventional F-dim and is defined as a matrix (F-matrix) satisfying the vector scaling equation. The parameter estimation procedure for the F-matrix is also discussed. Texture image analysing experiments were conducted to investigate effectiveness of the F-matrix as an image feature. The main results show that: (1) many texture images fit the F-matrix model well, (2) the F-matrix contains various information that characterize texture images, (3) the F-matrix was useful for texture classification, a 93.8% recognition rate is obtained for 65 samples in 13 categories.