i-manager's Journal on Image Processing (JIP)


Volume 6 Issue 1 January - March 2019

Research Paper

Optimized Video Compression Using Modified Intelligent Behaviour of Firefly Algorithm

H. A. Abdulkareem *, A. M. S. Tekanyi**, I. Yau***, K. A. Abu- Bilal****, H. Adamu*****
*-*****Ahmadu Bello University, Zaria, Nigeria.
Abdulkareem, H. A., Tekanyi, A. M. S., Yau, I., Abu-Bilal, K. A.,& Adamu, H.(2019). Optimized Video Compression Using Modified Intelligent Behaviour of Firefly Algorithm. i-manager's Journal on Image Processing, 6(1), 1-8.

Abstract

Transformation in mobile networks and multimedia communications make image and video compression important aspects of digital image processing. The main aim of image or video compression is to reduce the size of the image or video (redundancy) with little or no degradation of quality for an effective transmission and storage. This paper presents an optimized video compression using modified intelligent behavior of firefly algorithm. A total of six (four acquired and two benchmark) sample video data were used to implement the achieved technique. Frames were extracted from the video data and stored in the form of images in a buffer. Compression of the video frames was achieved by reducing the effect of pixel intensity with larger distance part. This was identified as one of the shortcomings with the Firefly Optimization Algorithm (FOA) method of image compression. In this paper, the impact of the modification was clearly shown using the Peak Signal to Noise Ratio (PSNR). The modification was achieved by including the root mean square in the standard equations of the FOA. In order to reduce the effect of pixel intensity with larger distance part, this was identified as one of the shortcomings. When the image samples were subjected to the (mFOA) compression technique, a same amount of improvement was achieved. Simulation results indicated that the mFOA technique outperformed the FOA method. The PSNR evaluation showed an improved reduction of frame size by 7.34%, 3.30%, 4.90%, and 5.75% for respective NAERLS1.avi, NAERLS2.avi, NTA1.avi, and NTA2.avi captured benchmark video frames and also 3.56% and 3.86% for respective video frames of Akiyo.avi and Forman.avi.

Research Paper

Skin Texture Recognition through Image Processing

Sukhdeep Sharma *, Aayushi Dubey**
*-**Department of Computer Science & Engineering, Manav Rachna International Institute of Research and Studies, Haryana, India.
Sharma, S., & Dubey, A. (2019). Skin Texture Recognition through Image Processing. i-manager's Journal on Image Processing, 6(1), 9-15.

Abstract

Image processing is done to get a better version of an image or an enhanced or filtered image. It is used to extract important information from the image. A raw image is converted into a digital image by undergoing various processes for which proper algorithms and mechanisms are defined. To further investigate on an image we can perform texture analysis on it. Texture recognition is also an important characteristic of image processing. Both texture recognition and image processing have a variety of uses in the medicine field. In this paper we have discussed about how we can detect the texture o f our skin and use it further to test the suitability of skin products that we use daily.

Research Paper

Primary Screening Technique for Detecting Breast Cancer

C. Naga Raju*, A Himabindhu**
* Associate Professor and Head, Department of Computer Science and Engineering, YSR Engineering College of Yogivemana University,Proddatur, Andhra Pradesh, India.
**Research Scholar, Department of Computer Science and Engineering, Ryalaseema University, Kurnool, Andhra Pradesh, India.
Raju, C. N., & Bindu, A. H.(2019). Primary Screening Technique for Detecting Breast Cancer. i-manager's Journal on Image Processing, 6(1),16-21.

Abstract

The breast cancer is absolutely life intimidating and dreadful disease. The primary screening of breast tumor is still under research because of some risk features such as gene, taking birth control fills, smoking, obesity and Age are playing vital role spreading the cancers. The malignant tumors that induct into the breast cells and eventually this tumor extends to the surrounding tissues. The proposed technique consists of four steps. Step1 is for digitized noises removal, step2 is for suppression of radio opaque artifacts, step3 is for Pectoral Muscle removal and step4 is for detecting location of cancer on breast for emphasizing the region of breast profile. To reveal the capability of this technique two separate digital mammograms are tested using GT( Ground Truth) mammograms for assessment of performance characteristics. The Experimental results indicate that the breast cancer regions are extracted truthfully in compliance to respective Ground Truth Images.

Research Paper

Performance Analysis of Copy-Move Forgery Detection Techniques

Suresh Gulivindala *, Srinivasa Rao Chanamallu **
*Department of Electronics & Communication Engineering, JNTUK University College of Engineering, Kakinada, Andhra pradesh, India.
**Department of Electronics & Communication Engineering, JNTUK University College of Engineering-Vizianagaram, Andhra pradesh, India.
Gulivindala, S., & Chanamallu, S. R. (2019). Performance Analysis of Copy-Move Forgery Detection Techniques. i-manager's Journal on Image Processing, 6(1), 22-30.

Abstract

Copy Move Forgery (CMF) is a manipulation process where a part of the image is copied and moved to another region in the same image. The advanced growth in technology and photo-editing software lead to the malicious manipulation of images. Distribution of such tampered images through high speed digital networks and social media is also increased which leads to the incredibility of the images and the underlying information. Hence, it is much demanded to develop, evaluate and propose CMF detection techniques. CMF detection can be achieved either by key-point approach or block-based approach. In this paper, performance of CMF detection and localization is evaluated.

Research Paper

Alzheimer’s Stage Classification using SVM Classifier using Brain MRI texture features

Shaik Basheera*, M. Satya Sai Ram**
*Department of ECE, Acharya Nagarjuna University College of Engineering, Acharya Nagarjuna University, Guntur, Andhra pradesh, India.
**Department of ECE, Chalapathi Institute of Engineering and Technology, LAM, Guntur, Andhra pradesh, India.
Basheera, S., & Ram, M. S. (2019). Alzheimer’s Stage Classification using SVM Classifier using Brain MRI texture features. i-manager's Journal on Image Processing, 6(1), 31-36.

Abstract

This paper deals with the Brain disorder caused due to dementia where the brain size gets effected and reduces its volume. Estimating the grade of Alzheimer’s is a challenging task. Spatial texture information is collected from T2 Weighted MRI images are used to perform classification and validation. We use 54 Brain MRI Slices, Gray Level Co-occurrence matrix is used to extract the attributes, and those are used to train the classifiers. On comparing Support Vector Machine (SVM) with Naïve Bayes classifier and KNN. SVM give the good classification accuracy of 98.1%. The classifiers classify the MRI into AD, MCI, and CN. 5 independent images collected from the Internet sources, by testing those images using SVM and correlate with clinical data of those images it achieves 100% accuracy.

Research Paper

Iris Recognition based on Optimized Orthogonal Wavelet and Local Tetra Pattern (OOWLTrP) using Neural Network

NuzhatF.Shaikh*
Professor, M. E. S. College of Engineering, Pune, India.
Shaikh, N. F. (2019). Iris Recognition based on Optimized Orthogonal Wavelet and Local Tetra Pattern (OOWLTrP) using Neural Network. i-manager's Journal on Image Processing, 6(1), 37-42.

Abstract

This paper proposes a novel feature descriptor using Optimized Orthogonal Wavelets and Local Tetra Patterns OOWLTrP. Texture features are extracted by using Local tetra pattern (LTrP) and wavelet features are extracted through optimized orthogonal wavelet. Eye image from various databases such as CASIA, MMUI and UBRIS are first preprocessed to remove salt and pepper noise. Later the iris is segmented from the rest of the eye image. Features are then extracted by using the proposed method which combines the goodness of local patterns and orthogonal wavelets. Feed Forward Back Propagation Neural Network (FFBNN) is used for classification of images. During training, the weights of FFBNN are optimized using the Adaptive Central Force Optimization (ACFO). The well trained FFBNN-ACFO is further used for classification of iris images. It has been observed that, there is considerable improvement in accuracy and validation time of the system. The increase in accuracy is due to the fact that LTrP extracts information from four directions as compared to LBP (Local Binary Pattern), LDP (Local Derivative Pattern) and LTP (Local Ternary Pattern). Coefficients of the orthogonal wavelet are optimization by genetic operators, this adds to the improvement in accuracy and reduction in validation time.