Accepted Papers

 

  • Effective Edge Detection: A Comparative Study
    Ahmed Saeed Mahdi1, Abdelmonim Artoli 2 and Awad Hag Ali Ahmed 3 ,1University of Neelain, Khartoum,Sudan ,2King Saud University, Saudi Arabia
    ABSTRACT
    Edge detection is the most important step in image segmentation and analysis. Statistical distributions are commonly used in this field. In this paper we present a comparative analysis between Beta, Log-Normal and Gamma distributions in terms of accuracy, complexity and computational performance. It was found that Log-Normal was best in terms of accuracy however more complex. Beta and gamma distributions were similar in both accuracy and performance for they both deal with asymmetric data. Anew Convolution from these distribution which benefits from there advantages and avoids there shortcomings is proposed and tested.
  • Using the Dual-Tree Complex Wavelet Transform for Improved Fabric Defect Detection
    Philibert Nsengiyumva1, Herman J. Vermaak2 and Nicolaas J. Luwes3, 1University of Rwanda, Rwanda,2Central University of Technology, South Africa,3Central University of Technology, South Africa
    ABSTRACT
    The dual-tree complex wavelet transform (DTCWT) solves the problems of shift-variance and low directional selectivity in two and higher dimensions found with the commonly-used discrete wavelet transform (DWT). I has been proposed for applications such as texture classification and content-based image retrieval.
    In this paper we evaluate the performance of the dual-tree complex wavelet transform for fabric defect detection. As experimental samples, we use the fabric images from TILDA, a textile texture database from the Workgroup on Texture Analysis of the German Research Council (DFG). We identify the mean energies of real and imaginary parts of complex wavelet coefficients taken separately as effective features for the purpose of fabric defect detection. We show that the use of the dual-tree complex wavelet transform yields greater performance as compared to the undecimated wavelet transform (UDWT) with a detection rate of 4.5% to 15.8% higher depending on the fabric type.
  • Identification of Patients with Obstructive Sleep Apnea Using the Entropy of the Histogram of the Continuous Wavelet Transform
    Abdulnasir Hossen, Sultan Qaboos University,Oman
    ABSTRACT
    A new method for identification of patients with obstructive sleep apnea (OSA) from normal controls is investigated in this paper using the entropy of the continuous wavelet transform. Two sets of data (train set and test set ) are used in this paper. Both sets are obtained from MIT databases. Each set consists of 20 OSA and 10 normal subjects. The identification factor between OSA and normal is obtained by finding the sum of the entropy of the histogram (with 10 levels) of the continuous wavelet transform (with 10 scales) of RRI data. The accuracy of classification approaches 88.66% on both training and test data.
  • Surveillance Video Based Robust Detection and Notification of Real Time Suspicious Activities in Indoor Scenarios
    Nithya Shree R, Rajeshwari Sah and Shreyank N Gowda, R.V.College of Engineering, India
    ABSTRACT
    Over recent years, surveillance camera is attracting attention due to its wide range of applications in suspicious activity detection. Current surveillance system focuses on analysing past incidents. This paper proposes an intelligent system for real-time monitoring with added functionality of anticipating the outcome through various Image processing techniques. As this is a sensitive matter, human decisions are given priority, still facilitating limited logical intervention of human resource. This framework detects risk in the area under surveillance. One such dangerous circumstance is implemented, like a person with a knife. Here the prediction is that in the firm places like ATM, Banks, Offices etc. a person possessing knife is unusual and likely to cause harmful activities like threatening, injuring and stabbing. The experiment demonstrates the effectiveness of the technique on training dataset collected from distinct environments. An interface is developed to notify concerned authority that boosts reliability and overall accuracy.

 

Copyright © SIGL 2016