MACHINE VISION ENT 273 Lecture 1 Ms HEMA

MACHINE VISION ENT 273 Lecture 1. Ms. HEMA C. R. Machine Vision System Components xxx Hema –ENT 273 – Lecture 1

Road Map u Image and Vision u Vision Systems u Components of an Machine Vision System [MVS] u Applications of vision systems u Advantages of MVS u Vision Optics u Frame Grabbers u Lighting and Illumination xxx Hema –ENT 273 – Lecture 1 2

Image and Vision u Image n Images are two-dimensional projections of the three-dimensional world u Vision n Vision is the most Complex of human senses, about a fourth of the brain’s volume is devoted to it. u Image Processing n Processing images to give new images u Computer Vision n Deals with what the images mean – aims to interpret images u Machine Vision n Apply vision and image processing u Vision System n A Vision System recovers useful information about a scene from its two dimensional projections xxx Hema –ENT 273 – Lecture 1 3

Machine Vision Systems u Characteristics n Ability to extract pertinent information from a background of irrelevant details n The capacity to learn from examples and apply to new situations n Ability to infer facts from incomplete information n Capability to generate self motivated goals and formulate plans for meeting these goals. xxx Hema –ENT 273 – Lecture 1 4

Components of a Machine Vision System n n n n xxx Hema –ENT 273 – Lecture 1 Input source l objects, scene, prints etc Optics l sensors, digital cameras Lighting l illumination levels A part sensor [optional] l to indicate presence of objects A frame grabber l stores images & interface PC platform [optional] Inspection software l Image processing algorithms Digital I/O l Display, Print, Interface 5

Vision System Portrayal xxx Hema –ENT 273 – Lecture 1 6

Operations to be performed by MVS u Describe images, objects and physical world n Mathematical models of image and objects and knowledge representation u Image Processing n Improves image for human and computer consumption, highlight / extract relevant feature u Segmentation n Extract features such a edge, regions, surfaces etc. u Pattern Recognition n Classify the images u Measurement Analysis n Measure features on the object u Image Understanding n Locate objects in the image, classify them and build 3 D models The Ultimate Aim of a Vision System is to recognize objects within a image xxx Hema –ENT 273 – Lecture 1 7

Applications of a Vision System u. Autonomous Vehicles u. The Human Face u. Industrial Inspection u. Medical Images u. Remote Sensing u. Surveillance u. Transport Reference: http: //www. bmva. ac. uk/apps/ xxx Hema –ENT 273 – Lecture 1 8

Autonomous Vehicles Transport Safety Aerial Navigation xxx Hema –ENT 273 – Lecture 1 9

The Human Face Head Modeling Face Recognition xxx Hema –ENT 273 – Lecture 1 10

Industrial Inspection Detecting Objects Machine parts xxx Hema –ENT 273 – Lecture 1 11

Medical Images Chromosomes Brain MRI xxx Hema –ENT 273 – Lecture 1 12

Remote Sensing Land Management Crop Classification xxx Hema –ENT 273 – Lecture 1 13

Surveillance Intruder Monitoring People Tracking xxx Hema –ENT 273 – Lecture 1 14

Transport Number Plate Traffic Control xxx Hema –ENT 273 – Lecture 1 Hema –ENT 496 – Lecture 1 15

Advantages of MVS in Industries u Cutting out defective goods u Making better use of raw materials u Cutting the cost of quality control u Enabling real-time process monitoring u Improving employment conditions xxx Hema –ENT 273 – Lecture 1 16

Vision Optics u Vision Systems n Stand alone n PC based u Smart Camera n Self contained [no pc req. ] l CCD image sensors l CMOS image sensors u Vision Sensors Neural Network-Based Zi. CAMs from JAI Pulnix n Integrated devices n No programming required n Between smart cams and vision systems u Digital Cameras n CCD image n CMOS image n Flash memory n Memory stick n Smart. Media cards A Cognex In-Sight n Removable [microdrives, CD, DVD] Vision Sensor xxx Hema –ENT 273 – Lecture 1 Compact Vision System from National Instruments 17

Imaging Sensors u Image sensors convert light into electric charge and process it into electronic signals u Image Sensors n Charge Coupled Device CCD All pixels are devoted to light capture l Output is uniform l High image quality l Used in cell phone cameras l n Complementary Metal Oxide Semiconductor CMOS l Pixels devoted to light capture are limited l Output is not uniform l High Image quality l Used in professional and industrial cameras xxx Hema –ENT 273 – Lecture 1 18

Frame Grabbers u. A frame grabber is a device to acquire [grab] and convert analog to digital images. Modern FG have many additional features like more storage, multiple camera links etc. xxx Hema –ENT 273 – Lecture 1 19

Frame Grabbers u A typical frame grabber consists of n a circuit to recover the horizontal and vertical synchronization pulses from the input signal; n An analog to digital converter n a colour decoder circuit, a function that can also be implemented in software n some memory for storing the acquired image (frame buffer) n a bus interface through which the main processor can control the acquisition and access the data. xxx Hema –ENT 273 – Lecture 1 20

Lighting u Correct lighting is the single most important design parameter in a vision system u Selection of a light source for a vision application is governed by three factors: n The type of features that must be captured by the vision system n The need for the part to be either moving or stationary when the image is captured. n The degree of visibility of the environment in which the image is captured. xxx Hema –ENT 273 – Lecture 1 21

Lighting Techniques u The three lighting techniques used in vision applications are: n Front lighting, n Back lighting n Structured lighting xxx Hema –ENT 273 – Lecture 1 22

Front Lighting Sources Ring Shape Lighting to detect loose caps Spot Lighting to check chip orientation in embossed tape Tube Lighting to detect stains on sheets Area type lighting to detect hole position in lead frames xxx Hema –ENT 273 – Lecture 1 23

Interesting Links Visit http: //www. machinevisiononline. org http: //www. eeng. dcu. e/~whelanp/proverbs. pdf to understand vision systems better References: http: //www. bmva. ac. uk/apps/ www. machinevisiononline. org http: //homepages. inf. ed. ac. uk/rbf/CVonline xxx Hema –ENT 273 – Lecture 1 24

Machine Vision End of Lecture 1 xxx Hema –ENT 273 – Lecture 1