Engineering design optimization problems are frequently found highly non-linear with complex constraints design variables. These constraints can be expressed in terms of range, structure similarity, signal-to-noise relationship, etc in computer vision. Image processing problems can be reformulated as optimization problems, especially bio-inspired problems. By selecting and designing proper object function, restoration, segmentation, thresholding etc of images could be done with superior results. Meta-heuristics are fit for those cases for optimal solutions. Cuckoos are recognized as charming...

Les problèmes d'optimisation de la conception technique sont souvent très non linéaires avec des variables de conception à contraintes complexes. Ces contraintes peuvent être exprimées en termes de portée, de similarité de structure, de rapport signal sur bruit, etc. en vision par ordinateur. Les problèmes de traitement d'image peuvent être reformulés en problèmes d'optimisation, en particulier les problèmes bio-inspirés. En sélectionnant et en concevant la fonction d'objet appropriée, la restauration, la segmentation, le seuillage, etc. des images pourraient être effectués...

Die kantenerhaltende Glättung ist ein sehr wichtiger Operator in der Bildverarbeitung und im Bereich der Computer Vision. Er unterdrückt das zufällige Rauschen, während die Kanten des Bildes intakt bleiben. Die bilaterale Filterung ist ein Werkzeug zur Glättung von Bildern unter Beibehaltung der Kanten durch eine nichtlineare Kombination von Bildwerten. Dabei wird die Intensität jedes Pixels durch einen gewichteten Durchschnitt der Intensitätswerte benachbarter Pixel ersetzt. Bilaterale Filterparameter sind der Bereichsparameter erhöht, d. h. SSIM, PSNR, NIQE, der bilaterale Filter...

Edge-Preserving Smoothing is a very important operator in the image processing and computer vision domain. It suppresses the random noise while keeping the edges in the image intact. Bilateral filtering is a tool which smooths images while preserving edges, by means of a nonlinear combination of image values. It replaces the intensity of each pixel with a weighted average of intensity values from nearby pixels. Bilateral filter parameters are the range parameter increases i.e. SSIM, PSNR, NIQE, the bilateral filter gradually approaches Gaussian convolution more closely because the range...

This research work investigates visibility degradation due to poor weather conditions and pro- poses methods to improve the colour, contrast, and visibility of digital images taken under such conditions. It is common knowledge that digital images need post processing not only for better human perception but also to be fed as input to computer vision application systems. Visibility improvement is a process of restoring the true appearance, i.e. recovering what the scene should have looked like on a clear day, by enhancing the colour, contrast and sharpening the details. Amongst the ample...