Keywords : Feature extraction

Straight Lines Detection Based on GA with New Modification on Baron's Method

Fawziya Mahmood Ramo; Nidhal H. Al-Assady; Khalil I. Al-Saif

AL-Rafidain Journal of Computer Sciences and Mathematics, 2010, Volume 7, Issue 3, Pages 61-72
DOI: 10.33899/csmj.2010.163911

In view of development on feature extraction in digital image based on feature straight line, GA has been used in this paper after hybrid it with Baron's Method to detect straight line, some developments are performed on the Baron's Method and we called it Genetic Developed Baron's Method (GDBM). The proposed method has been applied in many of sample. The experiments show that the proposed hybrid method in this paper is achieves high performance and it produce 90% detection rate.

Arabic Character Recognition Using Fractal Dimension

Khalil I. Alsaif; Karam Hatim Thanoon

AL-Rafidain Journal of Computer Sciences and Mathematics, 2009, Volume 6, Issue 1, Pages 169-178
DOI: 10.33899/csmj.2009.163776

In this work the concepts of the pattern recognition was used to recognize printed Arabic characters, and the Fractal geometric dimension method was used.
The input for the system is image, with bitmap format , then the image of character is recognized, and after that it is feeding to the OCR system. A feature space containing the values of the fractal dimension for the letters of Arabic was constructed. These features were used in the recognition phase. In this phase a comparison was made between the values in the feature space and the values of the letter inputs to be recognized, the comparison was done by the minimum Euclidian distance. Results of this work are 75% succeeded. And Matlab 6.5 is used to write the functions and subroutines for this work.

Latin Character Recognition using Neural Networks

Jamal S. Majeed; Aseel Ali; Amar S. Majeed

AL-Rafidain Journal of Computer Sciences and Mathematics, 2007, Volume 4, Issue 2, Pages 143-155
DOI: 10.33899/csmj.2007.164022

The aim of this work is to recognize the printed Latin's characters. In this work two methods for constructing the feature space are used. These methods are Variance and Fractal dimension methods, as a result they have real values for every character in the Latin's language, and from these values they constructed the feature space extractions for every character in the Latin's language. After that, these features are given to the Back Propagation network for recognizing the characters.
The result is a highest recognition for the characters is obtained, it is about 82.75% characters while the unrecognized characters are 17.25.

Speech Files Compression Based on Signal Feature

Khalil Alsaif; Saja J. Mohammed

AL-Rafidain Journal of Computer Sciences and Mathematics, 2007, Volume 4, Issue 1, Pages 57-79
DOI: 10.33899/csmj.2007.164003

In this research a new algorithm was suggested for compressing speech files added a new style for storing signals, The suggested idea of compression begins with recording the speech via the microphone, then starting the proposed processing steps as follows :

Removing silent period.
Select the number of resulted signal samples.
Segmenting the resulting signal to number of frames.
Applying one of the curves fitting algorithms and obtaining the coefficients for the mathematical representation.
Storing the results in a new file format with .ssc (Speech Signal Compression) extension.

While the decompression process consisted of the reversal compression process steps, the signal is reconstructed using curve fitting coefficients which were stored in the new file, followed by returning the selected sample, then returning the silent period to their original location and finally listening to the retrieved speech signal. When the proposed algorithm had been applied on the files with different speech contents, the compression ratio was approximately (16.283%), and the ratio of SEGSNR was approximately (25.195dB).