Digital Image Processing

Nahed Tawfik ¹, Mahmoud Fakhr ², Moawad I. Dessouky ³

Affiliations
1 Computers and Systems Department, The Electronics Research Institute, EG
2 Electronics and Communications Department, Cairo University, IN
3 Department of Electronics and Electrical Communications, Menoufia University, IN

Abstract

This paper presents a proposed method for corneal pattern classification using a Cepstral approach and SVMs. This approach based on the transformation of the corneal image to 1D signal, the feature extraction process and finally the classification process. MFCCs are one of the best feature extraction techniques used in 1D signal. This approach composed of two phases: a training phase and testing phase. In the first phase, a database of the corneal patterns is applied to obtain features from each corneal image, and then these features are used to train Support Vector Machines. In the second phase, features are extracted with the same steps in training phase from a set of new corneal images and finally a feature matching process is carried out. In this work, 1D signal used with time domain or in different discrete transform domains. The experimental results indicate that this technique achieves high classification rate up to about 100%.

Keywords

Corneal Images, MFCCs, Support Vector Machines (SVMs), DCT, DST, and DWT.

Full Text

     

Abdelrahman Selim ¹, Moawad I. Dessouky ², Mohiy M. Hadhoud ², Fathi E. Abd El-Samie

Affiliations
1 Menoufia University, Menouf, EG
2 Menoufia University, EG

Abstract

In this paper, we study the affect of using the spiral architecture instead of the square block decomposition in fractal compression. Comparisons with other systems like the conventional square and the simplified fractal compression systems are presented. A comparison with standard JPEG system is also introduced. We apply these types of fractal compression on a video sequence. We have found that in the case of using the spiral architecture in fractal compression, the produced or decoded image or video has a better visual quality than that produced with the conventional square system and the previously presented simplified system. We found also that all types of fractal compression are better than the JPEG standard.

Keywords

Image Compression, Fractal, Decomposition, Spiral.

Full Text

     

Nahed Tawfik ¹, Mahmoud Fakhr ¹, Al-Nuaimy Waleed ², Moawad I. Dessouky ³, Fathi E. Abd EI-Samie ⁴

Affiliations
1 Computers and Systems Department, The Electronic Research Institute, EG
2 Department of Electrical and Electronic Engineering, University of Liverpool, GB
3 Department of Electronics and Electrical Communications, Menoufia University, EG
4 Menoufia University, Menouf, EG

Abstract

In this paper, a new technique for feature extraction from corneal images is presented which can be applied for corneal pattern recognition. Most of the previous methods are based on segmentation of the corneal images which are restricted to certain planes. In this paper, a proposed method is applied on corneal images which have two main phases. Firstly, the 2-D images are lexicographically ordered to 1-D signals, and then the Mel Frequency Cepstral Coefficients (MFCCs) and polynomial coefficients are extracted from these 1-D signals or from their transforms. Secondly, the SVM is used to match the extracted features in the testing phase to those of the training phase. Experimental results show that the recognition rate for features extracted from Discrete Sine Transform DST and Discrete Cosine Transform (DCT) achieve better performance compared to other cases. The method in this paper is limited to feature extraction for pattern recognition and the automatic diagnosis case is left for future work.

Keywords

Corneal Images, Pattern Recognition, Mel Frequency Cepstral Coefficients (MFCCs), Polynomial Coefficients, Support Vector Machine (SVM), Discrete Cosine Transforms (DCT), Discrete Sine Transforms (DST), Discrete Wavelet Transforms (DWT).

Full Text

     

Zeinab F. Elsharkawy ¹, Safey A. Abdelwahab ¹, Moawad I. Dessouky ², Sayed M. Elaraby ³, Fathi E. Abd El-Samie ²

Affiliations
1 Engineering Department, Nuclear Research Center, Atomic Energy Authority, Cairo, EG
2 Department of Electronics and Electrical Communications, Menoufia University, Menouf, EG
3 Engineering Department, Nuclear Research Center, Atomic Energy Authority, Cairo, IN

Abstract

Scanner identification is the ability to discern the devices by which an image was scanned. In this paper, a new and robustness individual source scanner identification scheme is proposed. This scheme formulates a unique fingerprint for each scanner using traces of dust, dirt, scratches, and source imperfection pattern over scanner platen on scanned images. A single Support Vector machine (SVM) classifier is implemented and trained using correlation features of scanned images to classify different scanners brands and different models for the same scanner brand, and a 99.79% detection accuracy is obtained. In addition, the robustness of the used individual source scanner identification scheme on resized different resolutions is experimentally tested. The aging effect is also experimentally tested by re-applying the proposed algorithm on the scanned images after a continuous usage of the scanners under test for certain long periods. The experimental results using the proposed classifier for different scanner brands and different models for the same scanner brand approved the validity, efficiency, and robustness of the proposed scheme to match the scanned image to its unique source.

Keywords

Image Classification, Digital Image Forensics, Support Vector Machine.

Full Text

     

Amany F. Eldaoshy ¹, Moawad I. Dessouky ², Al-Dalil Sami ³, Fathi E. Abd El-Samie ²

Affiliations
1 Department of Electronics and Electrical Communications, Menoufia University, Menouf-32952, EG
2 Department of Electronics and Electrical Communications, Menoufia University, Menouf 32952, EG
3 Department of Electronics and Electrical Communications, Menoufia University, Menouf 32952, EG

Abstract

This paper presents a block-by-block SVD watermarking algorithm with variable block sizes. The paper makes a comparison between the traditional method of Liu and the proposed method. In the proposed approach, the original image is divided into blocks, and then the watermark is embedded in the singular values (SVs) of each block, separately. This segmentation and watermarking on a block-by-block basis makes the watermark more robust to the attacks such as noise, compression, cropping and other attacks as the results reveal. Watermark detection is implemented by extracting the watermark from the SVs of the watermarked blocks.  Extracting the watermark from one block at least is enough to ensure the existence of the watermark. Experimental results show that the proposed method with different block sizes is superior to the traditional method of Liu for embedding unencrypted watermarks.

Keywords

Image Processing, Cryptography, Watermarking, Singular Value Decomposition.

Full Text