RELATIVE POSITIONING OF STROKE-BASED CLUSTERING: A NEW APPROACH TO ONLINE HANDWRITTEN DEVANAGARI CHARACTER RECOGNITION

2012 ◽  
Vol 12 (02) ◽  
pp. 1250016 ◽  
Author(s):  
K. C. SANTOSH ◽  
CHOLWICH NATTEE ◽  
BART LAMIROY
Keyword(s):  
Character Recognition ◽  
Dynamic Time Warping ◽  
Large Range ◽  
Recognition Rate ◽  
Relative Positioning ◽  
Writing Style ◽  
Time Warping ◽  
New Approach ◽  
Tip Position ◽  
Dynamic Time

In this paper, we propose a new scheme for Devanagari natural handwritten character recognition. It is primarily based on spatial similarity-based stroke clustering. A feature of a stroke consists of a string of pen-tip positions and directions at every pen-tip position along the trajectory. It uses the dynamic time warping algorithm to align handwritten strokes with stored stroke templates and determine their similarity. Experiments are carried out with the help of 25 native writers and a recognition rate of approximately 95% is achieved. Our recognizer is robust to a large range of writing style and handles variation in the number of strokes, their order, shapes and sizes and similarities among classes.

A New Approach for an Efficient DTW in Face Detection through Eyes Localization

1970 ◽  
Vol 108 (2) ◽  
pp. 103-108
Author(s):  
S. Adwan ◽  
H. Arof
Keyword(s):  
Image Processing ◽  
Face Detection ◽  
Dynamic Time Warping ◽  
Weighting Scheme ◽  
Detection Accuracy ◽  
Processing Strategy ◽  
Time Warping ◽  
New Approach ◽  
Innovative Method ◽  
Dynamic Time

This paper presents an innovative method of face detection by supplementing dynamic time warping algorithm with proposed image processing strategy and weighting scheme. Using our proposed approach overcomes some of the shortcomings in applying dynamic time warping for face detection, hence improving the performance and detection accuracy. The results presented and discussed in this paper show the efficacy of our approach in using DTW for face detection. Ill. 16, bibl. 16, tabl. 1 (in English; abstracts in English and Lithuanian).http://dx.doi.org/10.5755/j01.eee.108.2.154

scholarly journals Recognition of Handwritten Characters using Deep Convolutional Neural Network

2019 ◽  
Vol 8 (6S4) ◽  
pp. 314-317
Keyword(s):  
Character Recognition ◽  
Optical Character Recognition ◽  
Recognition Rate ◽  
Writing Style ◽  
Handwritten Character Recognition ◽  
New Approach ◽  
Optical Character ◽  
Handwritten Character ◽  
Handwritten Recognition ◽  
Difficult Part

Handwritten character recognition (HCR) mainly entails optical character recognition. However, HCR involves in formatting and segmentation of the input. HCR is still an active area of research due to the fact that numerous verification in writing style, shape, size to individuals. The main difficult part of Indian handwritten recognition has overlapping between characters. These overlapping shaped characters are difficult to recognize that may lead to low recognition rate. These factors also increase the complexity of handwritten character recognition. This paper proposes a new approach to identify handwritten characters for Telugu language using Deep Learning (DL). The proposed work can be enhance the recognition rate of individual characters. The proposed approach recognizes with overall accuracy is 94%.

Stressed Speech Recognition Method Based on Difference Subspace Combined with Dynamic Time Warping

2012 ◽  
Vol 241-244 ◽  
pp. 1640-1646
Author(s):  
Cheng Guo Lv ◽  
Ru Bo Zhang ◽  
Pei Hua Li
Keyword(s):  
Dynamic Time Warping ◽  
Recognition Rate ◽  
Decision Criterion ◽  
Time Warping ◽  
Training Set ◽  
Recognition Method ◽  
Average Recognition Rate ◽  
Normal Environment ◽  
Acceleration Of Gravity ◽  
Dynamic Time

Speech under G-force which produced when speaker was under different acceleration of gravity was analyzed and researched, considered as principal part and stressed part to research. An isolated word recognition approach was proposed which combined difference subspace means with dynamic time warping technique. The method recognized speech under G-force by constructing a difference subspace to remove the stressed part. Dynamic time warping technique was adopted to make all feature vectors of one word in the training set have equal length, and a corresponding decision criterion was suggested. For a small vocabulary including 15 words, the method obtained the average recognition rate of 98.3%, which almost equal to the rate in normal environment. The method not only worked well in normal conditions but also had good performance for speech under G-force.

Follow That Tune-Adaptive Approach to DTW-based Query-by-Humming System

2015 ◽  
Vol 39 (4) ◽  
pp. 467-476 ◽  
Author(s):  
Bartłomiej Stasiak
Keyword(s):  
Dynamic Time Warping ◽  
Experimental Validation ◽  
Recognition Rate ◽  
Multimedia Databases ◽  
Time Warping ◽  
Query By Humming ◽  
Human Ability ◽  
Alignment Problem ◽  
Computationally Expensive ◽  
Dynamic Time

Abstract Dynamic Time Warping is a standard algorithm used for matching time series irrespective of local tempo variations. Its application in the context of Query-by-Humming interface to multimedia databases requires providing the transposition independence, which involves some additional, sometimes computationally expensive processing and may not guarantee the success, e.g., in the presence of a pitch trend or accidental key changes. The method of tune following, proposed in this paper, enables solving the pitch alignment problem in an adaptive way inspired by the human ability of ignoring typical errors occurring in sung melodies. The experimental validation performed on the database containing 4431 queries and over 5000 templates confirmed the enhancement introduced by the proposed algorithm in terms of the global recognition rate.

Speech Recognition for Endoscopic Automatic Positioning System

2012 ◽  
Vol 588-589 ◽  
pp. 1296-1299
Author(s):  
Ning Ma ◽  
Xiao Dong Chen ◽  
Ya Nan Li ◽  
Qing Yun Yin ◽  
Yi Wang ◽  
...  
Keyword(s):  
Speech Recognition ◽  
Embedded System ◽  
Dynamic Time Warping ◽  
Recognition Rate ◽  
Positioning System ◽  
Time Warping ◽  
Embedded Platform ◽  
Speech Recognition Engine ◽  
Dynamic Time ◽  
Automatic Positioning

A novel system for minimally invasive surgery is presented in this paper. The system utilized an Endoscopic Automatic Positioner (EAP) controlled by Speech Recognition Engine to implement the clamping and dynamically positioning of the laparoscope. The motion instructions of the EAP are transformed from voice commands of specific doctor recognized by an improved algorithm named Normalized Average- Dynamic Time Warping (NA-DTW). An embedded platform based on ARM is designed to run the NA-DTW on Windows CE operating system. 1250 groups of experiments from 10 individual speakers demonstrate the performance of DTW. Compared with traditional algorithms, the enhanced algorithm improves the recognition rate from 96.6% to 99.76% and shortens the time of calculation by 51%. The results demonstrate the enhanced algorithm being effective and can satisfy the real time requirement in embedded system.

Chinese Connected Word Speech Recognition Based on Derivative Dynamic Time Warping

2012 ◽  
Vol 542-543 ◽  
pp. 1324-1329
Author(s):  
Zhi Guo He ◽  
Ze Min Liu
Keyword(s):  
Speech Recognition ◽  
Computational Complexity ◽  
Dynamic Time Warping ◽  
Recognition Rate ◽  
Experimental Results ◽  
Time Warping ◽  
Test String ◽  
Cumulative Distance ◽  
Isolated Word ◽  
Dynamic Time

The algorithm of derivative dynamic time warping (DDTW) can overcome the shortcoming of dynamic time warping (DTW) and the computational complexity has not increased. In this paper, the algorithm of DDTW was applied to Chinese connected word speech recognition. For each isolated word, as an independent reference template and as basic recognition unit, there was an independent reference template to correspond; the matching between some word of the test string and a reference template was done by the DDTW, and the reference string which had the minimum cumulative distance was as output. The experimental results show that our method is obviously superior to all the methods based on DTW, and the recognition rate has reached 90%.

Sign in / Sign up

Export Citation Format

Share Document