Image Processing (Digital Halftoning, Compression, ML & DL Models)

function MT=DDMT() %DDMT- Dispersed  Dot Multitoning (MT) %Code by: Sankarasrinivasan S, %Multimedia Signal Processing Lab %NTUST, Taipei %Input Image im=imread('cameraman.tif'); %Reference %K. Chandu M. Stanich C. W. Wu B. Trager "Direct multi-bit search (DMS) screen algorithm" Proc. 19th IEEE Int. Conf. Image Process. pp. 817-820 Sep./Oct. 2012. %Dispersed Dot Multitone Screen generated using the referred paper %3-tone image require two screens DA=[ 141 189 208 102 251 174 217 165 198 252 204 129 180 150 217 249 174 245 69 223 145 191 134 245 107 137 188 224 244 209 160 199 219 97 231 123 197 235 207 83 213 231 173 158 142 177 239 147 188 163 215 139 241 101 171 227 152 133 238 197 251 193 136 228 89 252 152 203 82 221 116 249 190 211 167 222 99 214 85 207 156 196 227 177 247 98 205 162 94 243 113 185 165 234 182 242 236 185 135 212 148 194 233 182 218 195 226 206 247 125 202 140 210 159 240 172 225 107 142 239 139 175 155 96 194 173 221 176 130 217 ...

Digital Multi-toning is simply an up-gradation of conventional halftone technique. In halftone, a gray scale image is converted to a binary image useful for black and white printing (only use black cartridge). Similarly, digital color halftoning deals with production of printable format color images. Multitoning is a slight variation to halftone in which instead of binary image, it also use intermediate values of gray scale. function [Y]=CDM() %CDM- Clustered Dot Multitoning %Code by: Sankarasrinivasan S, %Multimedia Signal Processing Lab %NTUST, Taipei %Input Image im=imread('cameraman.tif'); %Reference % Y. F. Liu and J. M. Guo, "Clustered-Dot Screen Design for Digital Multitoning," in IEEE Transactions on Image Processing, vol. 25, no. 7, pp. 2971-2982, July 2016. %Clustered Dot Multitone Screen- Genereated using the referred paper. DA=[130 94 143 158 176 182 121 159 216 247 254 136 161 169 39 52 239 73 10 27 98 75 9 60 144 192 227 212 199 183...

To understand digital halftoning and ordered dithering please check this page http://imageprocessing-sankarsrin.blogspot.tw/2016/12/digital-half-toning-ordered-dithering.html Check the code Python: Ref:  https://github.com/SankarSrin/Digital-Halftoning MATLAB Code: %%%%%%%Function to generate Bayer's Halftone Patterns%%%%%%%%%% %%   Author: %%   Mr. Sankarasrinivasan %%   Research Fellow, Multimedia Signal Processing Lab, NTUST, Taiwan %%   Advisor: Prof. Jing Ming Guo %%   Dated: Apr, 2018 %%Details %%Bayers(im,met) im==> input image; %% met=1; Bayers Dispersed Dot && met=2; Bayers Clustered Dot function [HOD]=Bayers(im,met) im=im2double(im); [s1 s2]=size(im); if (met==1) %Bayers Dispersed Dot DA=[00 48 12 60 03 51 15 63;     32 16 44 28 35 19 47 31;     08 56 04 52 11 59 07 55;     40 24 36 20 43 27 39 23;     02 50 14 62 01 49 13 61;   ...

For introduction to Digital Halftoning refer to the following  http://imageprocessing-sankarsrin.blogspot.tw/2016/12/digital-half-toning-ordered-dithering.html http://imageprocessing-sankarsrin.blogspot.tw/2017/04/digital-half-toning-error-diffusion.html Dot Diffusion techniques is an improvisation of error diffusion with parallel processing feature of Ordered dithering. Basically it consist of two matrix such as Class and Diffusion Matrix. Class Matrix (CM) - Determines the order in which the pixels are processed in a block. (in other techniques raster or serpentine scan order is adopted). Note: All the blocks will be processed in parallel.  For example in Knuth Class matrix, '0' indicates the index of the first pixel to be processed and in further, pixels are processed with respect to positions of 1,2..63. Distribution Matrix - Provides the weight-age of error that has to be distributed to neighborhood pixels Abstract  Digital halftones...

For Full Article Refer  http://ieeexplore.ieee.org/abstract/document/6873260/ Perceptual image quality assessment (IQA) aims to use computational models to measure the image quality in consistent with subjective evaluations. Visual saliency (VS) has been widely studied by psychologists, neurobiologists, and computer scientists during the last decade to investigate, which areas of an image will attract the most attention of the human visual system. Intuitively, VS is closely related to IQA in that suprathreshold distortions can largely affect VS maps of images. With this consideration, we propose a simple but very effective full reference IQA method using VS. In our proposed IQA model, the role of VS is twofold. First, VS is used as a feature when computing the local quality map of the distorted image. Second, when pooling the quality score, VS is employed as a weighting function to reflect the importance of a local region. The proposed IQA index is called visual saliency-based...