A B C D E F G H I J K L M N O P Q R S T U V W X Y Z All
Maran, P.
- Watershed Algorithm Based Segmentation for Handwritten Text Identification
Authors
1 Department of Electronics and Communication Engineering, Velammal Engineering College, IN
2 Department of Electronics and Communication Engineering, Adhiparasakthi Engineering College, IN
3 Department of Electronics and Communication Engineering, SSN College of Engineering, IN
Source
ICTACT Journal on Image and Video Processing, Vol 4, No 3 (2014), Pagination: 767-772Abstract
In this paper we develop a system for writer identification which involves four processing steps like preprocessing, segmentation, feature extraction and writer identification using neural network. In the preprocessing phase the handwritten text is subjected to slant removal process for segmentation and feature extraction. After this step the text image enters into the process of noise removal and gray level conversion. The preprocessed image is further segmented by using morphological watershed algorithm, where the text lines are segmented into single words and then into single letters. The segmented image is feature extracted by Daubechies'5/3 integer wavelet transform to reduce training complexity [1, 6]. This process is lossless and reversible [10], [14]. These extracted features are given as input to our neural network for writer identification process and a target image is selected for each training process in the 2-layer neural network. With the several trained output data obtained from different target help in text identification. It is a multilingual text analysis which provides simple and efficient text segmentation.Keywords
Slant Correction, Morphological Watershed Algorithm, Daubechies’5/3 Integer-to-Integer Wavelet Transform, Neural Network.
- Numerical studies on convection in GTA Weld Pool
Authors
1 Department of Mechanical Engineering, Thiagarajar College of Engineering, Madurai, IN
2 Department of Mechanical Engineering, Indian Institute of Technology, Madras, Chennai, IN
Source
Indian Welding Journal, Vol 41, No 2 (2008), Pagination: 42-47Abstract
Weld pool convection strongly influences the behaviour of molten metal in the melt pool during fusion welding of metals. The temperature and velocity fields in the melt pool are largely affected by different driving forces causing weld pool convection. Variations in the heat input during welding have significant effects on the peak temperature, maximum velocity in the melt pool and weld bead geometry. Buoyancy, electromagnetic and surface tension are the major driving forces. In the present work, the effects of weld pool convection on weld bead geometry of stainless steel during Gas Tungsten Arc (GTA) welding have been studied for individual and combined driving forces. A two dimensional finite volume model has been used to simulate the welding process. The model uses a modified Gaussian heat distribution to provide the three dimensional effect of linear welding.
Keywords
Modeling, Weld Pool, Fluid Flow, GTA Welding, Stainless Steel, Weld Bead Geometry.- Microstructure and Hardness in GTA Welding of Stainless Steel
Authors
1 Department of Mechanical Engineering, Thiagarajar College of Engineering, Madurai-625015, IN
2 Department of Mechanical Engineering, Indian Institute of Technology, Madras, Chennai-600036, IN
Source
Indian Welding Journal, Vol 41, No 4 (2008), Pagination: 32-37Abstract
The experimental studies on microstructures and micro hardness in the weld metal show that the fluid flow pattern in the weld pool strongly influences the cooling rate, microstructure and hardness. The fluid pattern predicted is correlated with the experimental results for microstructures and hardness. The spatial variations of micro hardness show that the micro hardness values are higher in the low velocity fluid flow region and lower in the remaining region. The variations are due to different cooling rates, micro structures, sizes of the grains and delta ferrite content.