An MLP using Hough transform based fuzzy feature extraction for Bengali script recognition

Abstract

We define fuzzy sets on the Hough transform of character pattern pixels from which additional fuzzy sets are synthesized using t-norms. A multilayer perceptron trained with a number of linguistic set memberships derived from these t-norms can recognize characters of Bengali scripts by theirsimilarities to different fuzzy pattern classes.

Unified Software Development Model for Free/Open Source Software

“…Bengali (Bangla) is the national language of Bangladesh and one of the most popular languages of India [29]. There was no standardization of Bengali language writing system in computer.…”

“…So, operation of an Optical Character Recognition (OCR) on Bengali language is different from other languages [29]. We have developed a Bengali/Bangla OCR pre-processor.…”

Font-based persian character recognition using Simplified Fuzzy ARTMAP neural network improved by fuzzy sets and Particle Swarm Optimization

“…haracter recognition is one of the most popular areas of research in pattern recognition because of its immense application potential [1]. While character recognition has been extensively used as the basic application of different learning methods in machine learning literature for a few decades now, few researches have been done specifically for the Farsi language.…”

“…Types of features used in the literature include topological and geometrical, directional, mathematical, and structural features. Statistical analysis [4], structural and syntactic analysis [5], pattern matching [6], artificial neural networks [1], [7]- [8], fuzzy systems [1], [9] and genetic algorithms [10] are some of the computational techniques applied in this problem [5].Among them, An advantage of using neural networks is that a high computation rate can be achieved through their massive parallelism. It may be mentioned that human beings are more efficient than computers in handling complex recognition problems including character recognition from document images.…”

Recognition of an Indian script using multilayer perceptrons and fuzzy features

“…We had earlier reported a neuro-fuzzy technique for the recognition of Bengali characters [11]. We have extended this work by further fine-graining the classification process during recognition.…”