Generic Image Processing with Climb Laurent Senta – Didier Verna LRDE EPITA Research Lab April 30, 2012 lrde.epita.fr

Outline • Introduction • Using Climb • Developping Climb • Going Further • Conclusion 2

INTRODUCTION 3

Generic Image Processing 4 Images: Roland Levillain. Software Architecture for Generic Image Processing Tools I A H B D C E F G I A H B D C E F G segmentation

Genericity Purpose 5 Graphic: Laurent Najman. Point de vue d'un théoricien sur l'intérêt de la généricité pour le traitement d'images algorithms values type structures type segmentation graph 2dmatrix 3dmatrix bool grayscale rgb S x V x A combinations

Climb • Generic Image Processing library – Common Lisp • Inspired by Olena: C++, 10 years old – Dynamic vs Static – Usability, maintainability, performance,… • Still in beta 6 Compilers sbcl Dependencies iterate, lisp-magick, cl-heap, lisp-unit, cl-gtk2

Architecture 7 The different layers of the library Generic types Morphers Image Algorithms Chaining Operator GUI

USING CLIMB 8

Algorithms 9 original grayscale median filter Otsu threshold dilation erosion jitter

(setf img (image-load ‘images/lena128gray.jpg’)) (setf ot-img (otsu img)) (setf dil-img (dilation ot-img (4-connectivity))) (save dil-img ‘out/lena-dilated.png’) Chaining Operator Chaining algorithms with the $ operator 10 (save (erosion (otsu (image-load ‘images/lena128gray.jpg’)) (4-connectivity)) ‘out/lena-eroded.png’) image-load ‘images/lena128gray.jpg’ otsu dilation (4-connectivity) save ‘out/lena-eroded.png’ ($ (image-load ‘images/lena128gray.jpg’) (otsu) (dilation (4-connectivity)) (save ‘out/lena-eroded.png’))

dilation diff erosion otsu to-gray image-load Chaining Operator 11 More complex chaining

DEVELOPPING CLIMB 12

• matrix[x, y] → pixelvalue • graph.getNode(label) → nodevalue • model[x, y, z] → voxelvalue • matrix[x, y] → pixelvalue • graph.getNode(label) → nodevalue • model[x, y, z] → voxelvalue Image Definition 13 I A H B D C E F G Image access image(site) = value (setf (iref image site) value) Generalization: Lisp:

(value-inverse ) (iref image site) (setf (iref image site) ) Values and Sites 14 Values: Site: bool grayscale RGB x y I A H B D C E F G 2d-point label

Browsing Images 15 x y Site-set: (let ( ) (loop :for s := :while s )) (domain (image-domain image)) (site-set-next domain) … I A H B D C E F G

(let ( ) (loop :for s := :while s )) … (site-set-next neighbors) (neighbors (site-set-window window site)) Browsing Images 16 x y Site-set-window: I A H B D C E F G

Morphers 17 Transforming images with morphers Value morpher Structure morpher

GOING FURTHER 18

Properties 19 Adapting genericity algorithms values type structures type segmentation graph 2dmatrix 3dmatrix bool grayscale rgb support:regular

Value: GUI 20 Climb based interface I A H B D C E F G GTK-BOX Morpher Resize climb graph

Conclusion 21 Current status Image Processing Practitionner Algorithm implementor • Built-in algorithms • Composition tools • Chaining operator • Morphers • GUI • High-level domain model • Generic values • Generic structures • Generic implementations Hot topics Genericity Usability Performance

QUESTIONS ? Thanks for your attention, 22

Sources • Th. Géraud and R. Levillain. Semantics-driven genericity: A sequel to the static C++ object-oriented programming paradigm (SCOOP 2). • R. Levillain, Th. Géraud, and L. Najman. Why and how to design a generic and efficient image processing framework: The case of the Milena library. • N. Otsu. A threshold selection method from gray-level histograms. • P. Soille. Morphological Image Analysis: Principles and Applications • Roland Levillain. Software Architecture for Generic Image Processing Tools • Laurent Najman. Point de vue d'un théoricien sur l'intérêt de la généricité pour le traitement d'images 23