Course COMP 7116 Computer Vision Image Formation Geometric

Course : COMP 7116 – Computer Vision Image Formation: Geometric Camera Models Session 02 Prof. Dr. Widodo Budiharto 2018

Outline • • • Pinhole perspective Projective geometry Projection Matrix Adding a lens Focal Length, Aperture, and Depth of Fields

Camera and World Geometry

Pinhole Perspective f c f = focal length c = center of the camera Figure from Forsyth

Mapping from 3 D to 2 D

Projection • Its is just point of View!

Projective Geometry

Vanishing Points and Lines • Parallel lines in the world intersect in the image at “vanishing points”

Vanishing Points and Lines

Projection • World coordinate image coordinates

Common Image File Formats • • • GIF (Graphic Interchange Format) PNG (Portable Network Graphics) JPEG (Joint Photographic Experts Group) TIFF (Tagged Image File Format) PGM (Portable Gray Map) FITS (Flexible Image Transport System) T 0283 - Computer Vision 11

Color and grayscale image = ( (0. 3 * R) + (0. 59 * G) + (0. 11 * B) ) T 0283 - Computer Vision 12

Optical Flow • Optical flow or optic flow is the pattern of apparent motion of objects, surfaces, and edges in a visual scene caused by the relative motion between an observer (an eye or a camera) and the scene. T 0283 - Computer Vision 13

Display. Image. cpp // Menampilkan Image menggunakan cv. Load. Image #include "stdafx. h" #include <cv. h> #include <cxcore. h> #include <highgui. h> int _tmain(int argc, _TCHAR* argv[]) { Ipl. Image *img = cv. Load. Image("f: handsome. jpg"); cv. Named. Window("Open. CV", 1); cv. Show. Image("Open. CV", img); cv. Wait. Key(0); cv. Destroy. Window("Open. CV "); cv. Release. Image(&img); return 0; } T 0283 - Computer Vision 14

C++ 2. x Style //Display. Image 2. cpp #include "stdafx. h" #include <stdlib. h> #include <cv. hpp> #include <cxcore. hpp> #include <highgui. h> using namespace std; int _tmain(int argc, _TCHAR* argv[]) { cv: : Mat img=cv: : imread("f: /fruits. jpg"); cv: : imshow("Belajar Open. CV", img); //tampilkan cv: : wait. Key(); return EXIT_SUCCESS; } T 0283 - Computer Vision 15

Camera int main(int argc, char* argv[]) { Video. Capture cap(0); // open the video camera no. 0 if (!cap. is. Opened()) // if not success, exit program { cout << "Cannot open the video cam" << endl; return -1; } T 0283 - Computer Vision 16

Face Library • String face_cascade_name = "lbpcascade_frontalface. xml"; • String eyes_cascade_name = "haarcascade_eye_tree_eyeglasses. xml"; T 0283 - Computer Vision 17

sprintf(lbl. Object, "%s", prob_teta. Obstacle); cv: : put. Text(frame, "Distance Obs. 1: 53", cv. Point(10, 40), 1, 2, cv: : Scalar(255, 10, 0, 3)); Bina Nusantara University 18

Example • Displaying face detected from webcam with ellipse and rectangle usually need by robotics engineer, because it can be used to measure distance between camera and the object, the rectangle codes : cv. Circle( img, center, radius, color, 3, 8, 0 ); cv. Rectangle( img, cv. Point( r->x, r->y ), cv. Point( r->x + r->width, r->y + r>height ), CV_RGB( 0, 255, 0 ), 1, 8, 0 ); Bina Nusantara University 19

Stereo Vision Proyeksi titik pada kamera • 10/19/2021 Titik di kerangka image depan Iq(Iqx, Iqy) merupakan proyeksi titik pada kerangka kamera Cq(Cqx, Cqy, Cqz) ke kerangka image depan, dimana f melambangkan focal length. 20

Stereo Model • Titik 3 dimensi pada kerangka kamera stereo dapat direkonstruksi menggunakan proyeksi titik 2 dimensi dari kerangka image kanan dan kiri 10/19/2021 21

Depth Estimation • Menggunakan kamera stereo, diperoleh posisi wajah tiap moving obstacle sehingga dapat diperkirakan dihitung jarak moving obstacle tersebut. • Posisi titik 3 dimensi pada stereo camera : Catatan: 10/19/2021 (Purwanto D. , 2001) 22

Motion History

Tracking dengan motion history a) b) c) Deteksi orang lewat dan arah gerak menggunakan motion history 10/19/2021 24

Machine Vision

Exercise • Buat Program dengan webcam untuk deteksi wajah dan mata, serta wajah dibuat kotak.

References • Richard Szeliski. (2011). Computer Vision: Algorithms and applications. 01. Springer. Chapter 1 & 2. • David Forsyth and Jean Ponce. (2002). Computer Vision: A Modern Approach. 02. Prentice Hall. Chapter 1 & 2. • https: //cs. brown. edu/courses/cs 143/lectures/02. pdf