Page 14 - Dynamic Vision for Perception and Control of Motion

Contents xv

11.1.2 At Which Range Do Obstacles Have To Be Detected?.............. 333

11.1.3 How Can Obstacles Be Detected?.............................................. 334
11.2 Detecting and Tracking Stationary Obstacles........................................ 336
11.2.1 Odometry as an Essential Component of Dynamic Vision........ 336
11.2.2 Attention Focusing on Sets of Features...................................... 337
11.2.3 Monocular Range Estimation (Motion Stereo) .......................... 338
11.2.4 Experimental Results ................................................................. 342

11.3 Detecting and Tracking Moving Obstacles on Roads ........................... 343
11.3.1 Feature Sets for Visual Vehicle Detection ................................ 345
11.3.2 Hypothesis Generation and Initialization................................... 352
11.3.3 Recursive Estimation of Open Parameters and Relative State... 361

11.3.4 Experimental Results ................................................................. 366
11.3.5 Outlook on Object Recognition.................................................. 375

12 Sensor Requirements for Road Scenes ....................377
12.1 Structural Decomposition of the Vision Task ...................................... 378

12.1.1 Hardware Base........................................................................... 378
12.1.2 Functional Structure................................................................... 379
12.2 Vision under Conditions of Perturbation............................................... 380
12.2.1 Delay Time and High-frequency Perturbation ........................... 380

12.2.2 Visual Complexity and the Idea of Gestalt ................................ 382
12.3 Visual Range and Resolution Required for Road Traffic Applications. 383
12.3.1 Large Simultaneous Field of View............................................. 384
12.3.2 Multifocal Design ...................................................................... 384
12.3.3 View Fixation............................................................................. 385
12.3.4 Saccadic Control ........................................................................ 386

12.3.5 Stereovision................................................................................ 387
12.3.6 Total Range of Fields of View................................................... 388
12.3.7 High Dynamic Performance....................................................... 390
12.4 MarVEye as One of Many Possible Solutions ...................................... 391
12.5 Experimental Result in Saccadic Sign Recognition .............................. 392