A comparison study of processing and visualization in image analytics
Chanintorn Jittawiriyanukoon · ACCENTS Transactions on Image Processing and Computer Vision · 2020
The authors also point out that various practical applications like animation involve with recognition technology.Support Vector Machine (SVM) algorithm is obtained to categorize the action and appearance (emotion) out of facial pattern recognition.The smoothing process is to ease unwanted elements (noise) and enhance features of the image graphic.Both non-linear [5] and linear [6] algorithms are functional for smoothing digital images.To smooth the image helps comfort multiple processes in image analytics.Linear smoothers are techniques that involve with a pixel intensity, which gives the linear sum amount in a sliding window.It is correspondent to the Fourier transform by multiplying and integrating for each image.Let a linear model denote as yi = aix, where x identifies the unknown variables with a (j x 1) matrix, ai represents an (i x j) 2D matrix regarding input parameters from original image, and yi contains the (i x 1) matrix of observed output.Regarding this recursive computation, i input parameters are employed with j unknown variables and each iteration results i observed outputs.