354 | P a g e
DESIGNING AND IMPLEMENTING AN
OPTIMIZATION ALGORITHM IN MATLAB TO ENHANCE THE QUALITY OF A DIGITAL IMAGE
Komal Rani
1, Kapil Kaswan
21
M.Tech (CSE), Chaudhary Devi Lal University, Sirsa, (India)
2
Assistant Professor, Deptt. of CSA, Chaudhary Devi Lal University, Sirsa, (India)
ABSTRACT
Resolution of a digital image is an important aspect in most of the image and video-processing applications, such as enhancement of resolution on an image, feature extraction in images, and resolution enhancement of satellite image.
Interpolation is the method which is used in processing of images to increase pixels quantity in a digital image.
Visualization and image processing are two main methods of producing images either through signal processing or from a diagram. Both would result to an image either printed or through the monitor. Both also have their own uses and applications and it is best to learn the difference between the two.
Keywords: Domain, Image Resolution, Pixel Intensity, Quality Improvement.
I. INTRODUCTION
Resolution is designated as significant holding for images. Resolution is mainly an underlined expression of any image. Gratifying quality images are required in every field. Here we propose new improved algorithm having extension in scaling and then the algorithm is applied on image mingled with text for producing a pleasing quality of source image.
There are four key areas in image processing
Image Segmentation
Image Registration
Image Visualization
Image Simulation
II. PROPOSED ALGORITHM
The Resolution Enhancement process has been implemented by using the following steps:
1. Select the image and taken it as input.
2. Read image pixel by pixel in MATLAB Image Toolbox.
3. For each pixel, generate enhanced image 4. Normalize Each value to [0,255]
5. Calculate intensity values and generate resolution enhanced image.
6. Generate RMSE & PSNR values to identify the results accuracy
7. Generate the Histograms of the images before and after resolution enhancement
355 | P a g e 2.1 Procedure to convert RGB Image to HSI Image:
function hsi = rgb2hsi(rgb)
rgb = im2double(rgb); r = rgb(:, :, 1); g = rgb(:, :, 2); b = rgb(:, :, 3);
num = 0.5*((r - g) + (r - b));
den = sqrt((r - g).^2 + (r - b).*(g - b));
theta = acos(num./(den + eps));
H = theta;
H(b > g) = 2*pi - H(b > g);
H = H/(2*pi);
num = min(min(r, g), b);
den = r + g + b;
den(den == 0) = eps;
S = 1 - 3.* num./den;
H(S == 0) = 0;
I = (r + g + b)/3;
hsi = cat(3, H, S, I);
function rgb = hsi2rgb(hsi)
H = hsi(:, :, 1) * 2 * pi; S = hsi(:, :, 2); I = hsi(:, :, 3);
R = zeros(size(hsi, 1), size(hsi, 2));
G = zeros(size(hsi, 1), size(hsi, 2));
B = zeros(size(hsi, 1), size(hsi, 2));
356 | P a g e idx = find( (0 <= H) & (H < 2*pi/3));
B(idx) = I(idx) .* (1 - S(idx));
R(idx) = I(idx) .* (1 + S(idx) .* cos(H(idx)) ./ ...
cos(pi/3 - H(idx)));
G(idx) = 3*I(idx) - (R(idx) + B(idx));
idx = find( (2*pi/3 <= H) & (H < 4*pi/3) );
R(idx) = I(idx) .* (1 - S(idx));
G(idx) = I(idx) .* (1 + S(idx) .* cos(H(idx) - 2*pi/3) ./ ...
cos(pi - H(idx)));
B(idx) = 3*I(idx) - (R(idx) + G(idx));
idx = find( (4*pi/3 <= H) & (H <= 2*pi));
G(idx) = I(idx) .* (1 - S(idx));
B(idx) = I(idx) .* (1 + S(idx) .* cos(H(idx) - 4*pi/3) ./ ...
cos(5*pi/3 - H(idx)));
R(idx) = 3*I(idx) - (G(idx) + B(idx));
rgb = cat(3, R, G, B);
rgb = max(min(rgb, 1), 0);
III. RESULTS & DISCUSSIONS
Figure 1: Input Image 1 Figure 2: Resolution Enhanced Image 1
Figure 3: Input Image Histogram 1 Figure 4: Output Image Histogram 1
357 | P a g e Figure 5: Input Image 6 Figure 6: Resolution Enhanced Image 6
Figure 7: Input Image Histogram 6 Figure 8: Output Image Histogram 6
Table 1: PSNR & RMSE Values for Existing & Proposed Algorithm
Image No. Existing PSNR Proposed PSNR Existing RMSE Proposed RMSE
I1 26.4840917 26.6800736 12.13 11.86
I2 27.6543720 30.7262473 10.61 7.45
I3 27.9528129 33.7395444 10.25 5.26
I4 30.3652587 39.4606871 7.76 2.72
I5 29.3578015 51.8112473 8.72 0.66
I6 28.3780565 61.4146433 9.76 0.22
Figure 9: Comparison of PSNR & RMSE Values for Existing & Proposed Algorithm 0
20 40 60 80
I1 I2 I3 I4 I5 I6
Old PSNR New PSNR
0 5 10 15
I1 I2 I3 I4 I5 I6
Old RMSE New RMSE
358 | P a g e
IV. CONCLUSION & FUTURE SCOPE
We have seen the significance and requirement of the image resolution enhancement. We have studied the Discrete Wavelet Transform and Stationary Wavelet Transform methods of image resolution enhancement. We have analyzed the problem associated with these methods and proposed a different technique for the same. Further we have implemented this technique and finally we analyzed the results of proposed method on the basis of the various performance parameters i.e. PSNR and RMSE. We have also generated Histograms of the images before and after resolution enhancement. All the results are shown in the tabular form. From the results, we found that our proposed technique shows significantly better results as compared to various existing techniques. We may utilize this algorithm for the enhancement of various poor resolution images. In the future, we may try to get some technique which provides more improvement in the PSNR and RMSE, to get the super resolution images. The proposed work can also be extended in future to improve quality of images of moving pictures i.e. videos and also the moving images combined with audio files. This work can also be extended to improve quality of various animated images of different formats. Images and videos can be enhanced in their quality at different application softwares and various operating systems platforms.
REFERENCES
[1] Steven McDonagh, Benjamin Hou, “CONTEXT-SENSITIVE SUPER-RESOLUTION FOR FAST FETAL MAGNETIC RESONANCE IMAGING”, Biomedical Image Analysis Group, Imperial College London, 4 March, 2017.
[2] Yerramsetty Prabhakar, K. Khamar Taj, “INTERNATIONAL JOURNAL OF INNOVATIVE TECHNOLOGIES”, Vol.04, Issue.03, March-2016, Pages: 0391-0394.
[3] Mohit, Krishan Kumar, “RESOLUTION ENRICHMENT TECHNIQUES FOR IMAGE INCORPORATED WITH TEXT”, International Journal of Advanced Research in Computer and Communication Engineering, Vol. 4, Issue 8, August 2015.
[4] Pravin R. Dabhi, Prof. S G.Bari, “IMAGE ENHANCEMENT”, Journal of Emerging Technologies and Innovative Research, Volume 2, Issue 3, March 2015.
[5] Tarun Gulati, Kapil Gupta, Dushyant Gupta, “BETTER PERCEPTION OF LOW RESOLUTION IMAGES USING WAVELET INTERPOLATION TECHNIQUES”, International Journal of Electrical, Computer, Energetic, Electronic and Communication Engineering Vol:9, No:4, 2015.
[6] R. Sowmya, Golla Ramesh, Mohammed Javeed, “RESOLUTION ENHANCEMENT OF IMAGES WITH INTERPOLATION AND DWT-SWT WAVELET DOMAIN COMPONENTS”, International Journal of Science, Engineering and Technology Research, Vol.04, Issue 17, June-2015, Pages: 3158-3163.
[7] Rebika Rai, Ratika Pradhan2, M.K. Ghose, “ANT BASED SWARM COMPUTING TECHNIQUES FOR EDGE DETECTION OF IMAGES”, International Journal of Emerging Technology and Advanced Engineering, Volume 3, Issue 4, April 2013.
