Segregated Handwritten Character Recognition using GLCM features

Abstract

Handwritten document recognition is an area of pattern recognition that has been showing impressive performance in the machine printed text. Handwritten document recognition is an intricate task to various writing styles of individual person. The system first identifies the contour in a handwritten document for segmentation and features are extracted from the segmented character. This paper uses GLCM(Gray Level Co-occurrence Matrix) for character recognition. Features of a character has been computed based on calculating the pairs of pixel with specific values and specified spatial relationship occurrence in an image. First order and second order textures are used to measure the intensity of the original pixels. Data were collected from different persons, and the system is trained using SVM with various writing styles. The proposed system achieves a maximum recognition accuracy of 95.2% with training and testing data using GLCM as features and SVM with RBF kernel function.

Pengenalan Pola Tulisan Tangan Aksara Bima menggunakan Ciri Tekstur dan KNN

“…Salah satu metode yang cukup handal dalam melakukan ekstraksi ciri tekstur adalah Gray Level Co-occurance Matrix (GLCM). GLCM beberapa kali digunakan dalam mengektraksi ciri pada tulisan tangan aksara, dan memberikan performa yang cukup baik yakni berkisar antara 80 sampai 95% [18], [21].…”

“…9 ciri GLCM dicari menggunakan 4 sudut, sehingga didapat total ciri yang akan digunakan adalah 36 ciri. berdasarkan pengujian, hasil yang didapatkan menunjukkan kemampuan pengenaln pola yang cukup baik yaitu berkisar antara 80.76% sampai dengan 95.43% [21] Singla pada tahun 2014 menulis tentang penggunaan 4 ciri GLCM meliputi contrast correlation, Energy dan Homogeneity pada Jaringan syaraf tiruan untuk mengenali pola aksara Devanagari. Pengujiannya memanfaatkan 10fold validation, dengan pembagian dari 30 data 20:10 untuk data latih berbanding data uji mendapatkan hasil yang baik [19].…”

3D Motion Trajectories Recognition using Optimal Set of Geometric Primitives, Angular and Statistical Features

“…The features which are related to spatial relationship of pixels are characterized as statistical features. The relative positions of these gray levels are extracted from the second order statistical features [15,16]. The co-occurrence of gray level structure can be represented by a matrix related to distance  and orientation angle , that is [10,13,15,16] given by equation 6.…”

“…Total energy: It is used to measure the regularity of an arbitrary pair of pixels [10,13,15,16] given by equation 8.…”

“…It is used to extract second order statistical information of neighboring points or pixels present in image. We have studied total three statistical features namely contrast[10,13,15,16] of dimension four, homogeneity[10,13,15,16] of dimension four and total energy[10,13,15,16] of dimension four. In this work, three statistical features of total dimension 12 are calculated and used for recognition.…”