KMITL

Permanent URI for this communityhttps://dspace.kmitl.ac.th/handle/123456789/1

Browse

Settings

Access type: UnknownSubject: search.filters.subject.Pattern Recognition

Search Results

Now showing 1 - 2 of 2
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Publication,
    Adaptive Histogram of Oriented Gradient for Printed Thai Character Recognition
    (2014-01-01)
    Woraratpanya, Kuntpong
    ;
    Titijaroonroj, Taravichet
    A similarity of printed Thai characters is a grand challenge of optical character recognition (OCR), especially in case of a variety of font types, sizes, and styles. This paper proposes an effective feature extraction, adaptive histogram of oriented gradient (AHOG), for overcoming the character similarity. The proposed method improves the conventional histogram of oriented gradient (HOG) in two principal phases, which are (i) adaptive partition for gradient images and (ii) adaptive binning for oriented histograms. The former is implemented with quadtree partition based on gradient image variance so as to provide for an effective local feature extraction. The later is implemented with non-uniform mapping technique, so that the AHOG descriptor can be constructed with minimal errors. Based on 59,408 single character images equally divided into training and testing samples, the experimental results show that the AHOG method outperforms the conventional HOG and state-of-the-art methods, including scale space histogram of oriented gradient (SSHOG), pyramid histogram of oriented gradient (PHOG), multilevel histogram of oriented gradient (MHOG), and HOG column encoding algorithm (HOG-Column). © Springer International Publishing Switzerland 2014.
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Publication,
    Experimental results of using rough sets for printed Thai characters recognition
    (2001-12-01)
    Mitatha, S.
    ;
    Dejharn, K.
    ;
    Chevasuvit, F.
    ;
    Chankuang, B.
    ;
    Kasemsiri, W.
    This paper proposes the experimental results of using a Rough sets for the recognition of printed Thai characters. In our experiment, we segment each character into 32 pieces sized 4×4 pixels, and then find the distribution of pixels that match a value of "1" (black dot) in each section Following this, we use the resulting 32 values as the attributes for each given object. Afterwards, we create 3 sets of decision making rules from 3 different training sets and use those 3 set of rules to classify each member of the unknown set. This set is composed of 42 Thai characters, excluding the two that are very rarely used, with 7 fonts and 7 sizes, for a total of 2058. The results are 46.20%, 63.15%, 73.12% for the first set of rules, the second set of rules and the third set of rules respectively. And the results when apply the set of rule to the unknown which related to each set of rules's training sets are 100% for all three set of rules.