Team C Krishna Pal (200601005) Jitesh Prajapat(200601036) - PowerPoint PPT Presentation

Team C Krishna Pal (200601005) Jitesh Prajapat(200601036)

Presentation Flow:- • What is Computer Vision? • Examples of Computer Vision. • Tools of Computer Vision. • Related Disciplines. • Challenges in Embedded Systems. • Imaging Techniques. • Research and Application Areas.

What is Computer Vision? • Computer Vision is the science and technology of machines that see. • From Science perspective , computer vision is concerned with the theory for building artificial systems that obtain information from images. • From Technological perspective , computer vision seeks to apply the theories and models of computer vision to the construction of computer vision systems.

Examples of Computer Vision:-  Controlling processes (e.g. an industrial robot or an autonomous vehicle).  Detecting events (e.g. for visual surveillance or people counting).  Organizing information (e.g. for indexing databases of images and image sequences).  Modeling objects or environments (e.g. industrial inspection, medical image analysis or topographical modeling).  Interaction (e.g. as the input to a device for computer- human interaction).

Tools of Computer Vision:- • Hardware:- o For Acquiring and storing digital images in computer. o Processing the images. o Communicating results to users or other automated systems. • Software:- o For efficient algorithms.

Related Disciplines:-  Artificial Intelligence.  Machine Vision.  Image processing.  Pattern Recognition.

Related Disciplines:-  Artificial Intelligence :- o Deals with autonomous planning or deliberation for systems which can perform mechanical actions such as moving a robot through some environment. o This type of processing typically needs input data provided by a computer vision system, acting as a vision sensor and providing high-level information about the environment and the robot.

Related Disciplines:-  Machine Vision:- o Tends to focus on applications, mainly in industry, e.g., vision based autonomous robots and systems for vision based inspection or measurement. o External conditions such as lighting can be and are often more controlled in machine vision than they are in general computer vision.

Related Disciplines:-  Image processing:- o It concerns image properties and image to image transformation. o Image processing tends to focus on 2D images whereas Computer vision tends to focus on 3D images.

Related Disciplines:-  Pattern Recognition:- o It is a field which uses various methods to extract information from signals in general, mainly based on statistical approaches. o For recognizing and classifying objects using digital images.

Related Disciplines:-

Challenges in Embedded Systems:- • Embedded Computer Vision Systems have to balance their huge computational and communication demands with the stringent size , power, and memory resource constraints of embedded platforms. • These systems typically require fast processors , lots of RAM for storing images, and large amounts of power while images are being processed. • The end result is a fairly complicated, expensive system that is out of reach for many developers who have moderate image processing requirements.

Imaging Techniques:- • Normal Imaging – Capturing with a Digital Camera. Images stored in the form matrices of numbers. • IR Imaging – Infra red Imaging is used extensively for both military and civilian purposes. Involves tracking of the heat signature. • Electric Field Imaging – Imaging by tracking changes in the electrostatic field generated by the system. Used in nature by small electric fish for navigation.

Imaging Techniques:- Electric Field Imaging:-  Electric-field imaging starts with the electric field generated by a voltage potential between two conductors.  Works by measuring changes electric field in proximity of an object. The Motorola MC33794, together with a microcontroller, simplifies electric-field imaging. The chip supports up to nine electrodes and has built-in watchdog and power-on-reset timers.

Research and Application:- • Industrial inspection and quality control. • Surveillance and security. • Face recognition. • Road monitoring. • Autonomous vehicles. • Robotic systems. • Space applications. • Military application. • Medical Application.

Research and Application:- Surveillance and security:- o Recent advances in imaging and embedded computing technology enable wide scale computer vision systems that incorporate arrays of low-cost, portable imagers connected through wireless networks. o Video surveillance is a particularly important driving application area.

Research and Application:- Example of Surveillance Systems:- Detecting Illegal Parking • Background Object :- are the stationary and persistent objects in a scene. • Foreground Object :- objects which keep moving. • Midground Object :-moving foreground objects that stop and remain stationary for a given period of time before either moving again or migrating into the scene's background.

Research and Application:- Example of Surveillance Systems:-

Research and Application:- Example of Surveillance Systems:-

Research and Application:- • Face recognition:- o A facial recognition system is a computer application for automatically identifying or verifying a person from a digital image or a video frame from a video source. o One of the ways to do this is by comparing selected facial features from the image and a facial database.

Research and Application:- • Face recognition:- o This technique uses 3-D sensors to capture information about the shape of a face. o This information is then used to identify distinctive features on the surface of a face, such as the contour of the eye sockets, nose, and chin. o One advantage of 3-D facial recognition is that it is not affected by changes in lighting like other techniques. It can also identify a face from a range of viewing angles, including a profile view.

Research and Application:- • Face pattern capturing:-

Research and Application:- • Face detection:-

Research and Application:- • Face recognition:-

Research and Application:-  Robotic Application:-  Commonly, sensor technology is used for applications involving Robotic Vision.  Computer Vision is fast become one of the most important necessities of modern robotic systems.  It is specifically useful where embedded system has to make decisions on its own in real time.

Research and Application:-  Example of Robotic Application:- Mars path finder o It had a small robot called the Sojourner which used computer vision techniques to maneuver itself on the surface of the planet.

Research and Application:-  Autonomous Vehicle:- Driverless car o The driverless car is an autonomous vehicle that can drive itself from one point to another without assistance from a driver. o Features of Driverless car-  Anti-Lock Brakes:- With anti-lock brakes, the system does the pumping for the driver -- and does it better than the driver. The system can read the wheels and knows when they are about to lock and react faster and with a more appropriate response than a driver could.  Traction or Stability control:- Stability and traction control are systems that can detect when a car might go into an out-of-control skid or roll over and work to prevent that from happening. The systems are constantly reading the car's direction, speed and how well each wheel is connecting to the road.

Research and Application:-  Autonomous Vehicle:- Driverless car

Research and Application:-  Autonomous Vehicle:- Driverless car  Pre-safe Systems:-they can anticipate crashes and prepare the car to keep the occupants safe an alarm might go off as the driver nears the stopped car. At the same time, the pre-safe system might start priming the brakes so that just touching the pedal will apply their full force. While all that's going on, the car will start reducing engine power, which will slow the car and reduce the severity of the crash.  Cruise Control:- Using radar sensors on the front of the car, adaptive cruise control can tell when an object is in front of it and, if the object is moving, how fast it's moving. When cruise control is set, adaptive cruise control will maintain a constant speed, but will also maintain a set distance between it and the car in front of it.

Research and Application:-  Autonomous Vehicle:- Driverless car Cruise control System

Research and Application:-  Medical Applications:- o This area is characterized by the extraction of information from image data for the purpose of making a medical diagnosis of a patient. o Generally, image data is in the form of microscopy images, X-ray images, angiography images, ultrasonic images, and tomography images. o An example of information which can be extracted from such image data is detection of tumours, arteriosclerosis or other malign changes. It can also be measurements of organ dimensions, blood flow, etc.