9/10/2002 CS 461, Copyright G.D. HagerComputer Vision, Lecture 1http://www.ugrad.cs.jhu.edu/~cs461Professor Hagerhttp://www.cs.jhu.edu/~hager9/10/2002 CS 461, Copyright G.D. HagerOutline for Today• Outline and Organization of the course• What is Computer Vision• Some Applications of Computer Vision9/10/2002 CS 461, Copyright G.D. HagerWhat Information is in Images?9/10/2002 CS 461, Copyright G.D. HagerWhat Information is in Images?9/10/2002 CS 461, Copyright G.D. HagerWhat is Computer Vision?• Trucco and Verri– computing properties of the 3D world from one or more digital image• Stockman and Shapiro– To make useful decisions about real physical objects and scenes based on sensed images• Ballard and Brown– The construction of explicit, meaningful description of physicalobjects from images9/10/2002 CS 461, Copyright G.D. HagerSome Related Terms• Image Processing: the study of the properties of operators that produce images from other images– we will touch on image filtering and related operators from image processing• Machine Vision: a somewhat outdated term which now tends to refer to industrial vision applications where (usually) a singlecamera is used to solve a structured inspection task– the “reverse CAD” model• Pattern Recognition: typically refers to the recognition of structures in 2D images (usually without reference to any underlying 3D information).• Photogrammetry: the science of measurement though non-contact sensing, e.g. terrain maps from satellite images. Usually is more focused on accuracy issues than interpretation.9/10/2002 CS 461, Copyright G.D. HagerPixelBinary1 bitGrey1 byteColor3 bytesOur Data StructureEach pixel is a measure of the brightness (intensity of light)that falls on an area of an sensor (typically a CCD chip)9/10/2002 CS 461, Copyright G.D. HagerProblems of Computer Vision: ModelingWhat are the physical and geometric processes thatgovern (digital) imaging?9/10/2002 CS 461, Copyright G.D. HagerProblems of Computer Vision: ModelingWhat are the physical and geometric processes thatgovern (digital) imaging?9/10/2002 CS 461, Copyright G.D. HagerGeneral RulesIf you can’t understand (i.e. model)the forward process, you will havea hard time solving the inverse!A related point: the best way totest vision algorithms is always toimplement the forward modelto test the (inverse) solution.9/10/2002 CS 461, Copyright G.D. HagerComputer Vision vs. Graphics• Computer Graphics– Produce “plausible” images– You choose the models, conditions, imaging parameters, etc.• Computer Vision– Given real images with noise, sampling artifacts …– Estimate physically quantities– Ill-posed ---- what is the minimum world knowledge we need?Is Vision the “Inverse” of Graphics?9/10/2002 CS 461, Copyright G.D. HagerImageFilterResultProblems of Computer Vision: Feature ExtractionWhat are the “informative” areas of an image and how do we detect them?9/10/2002 CS 461, Copyright G.D. HagerFilter kernels that are larger see effects at coarserscales -- the filter on the left responds to the zebra’swhiskers, that on the right to its stripesProblems of Computer Vision: Feature Extraction9/10/2002 CS 461, Copyright G.D. HagerThresholding suppresses “non-feature” areasof the imageProblems of Computer Vision: Feature Extraction9/10/2002 CS 461, Copyright G.D. HagerComputer Vision vs. Image Processing• Image Processing– Mostly concerned with image-to-image transformations• Filtering• Enhancement• Compression• Computer Vision– Concerned with how images reflect the 3D world• Filtering for feature extraction• Enhancement for recognition/detection• Compression that preserves geometric information in images9/10/2002 CS 461, Copyright G.D. HagerProblems of Computer Vision: Segmentation and GroupingWhat portionsof an image pertainto one another andto relevant physicalphenomena?9/10/2002 CS 461, Copyright G.D. HagerComputer Vision vs. Human VisionWhat is the rightsegmentation?To us it seemsobvious …9/10/2002 CS 461, Copyright G.D. HagerObjective Reality vs. Subjective RealityMetric Geometry vs. “Shape”SymmetryThe color orange......9/10/2002 CS 461, Copyright G.D. HagerIllusions: What Do They Tell Us?9/10/2002 CS 461, Copyright G.D. HagerIllusions: What Do They Tell Us?9/10/2002 CS 461, Copyright G.D. HagerIllusions: What Do They Tell Us?9/10/2002 CS 461, Copyright G.D. HagerIllusions: What Do They Tell Us?9/10/2002 CS 461, Copyright G.D. HagerIllusions: What Do They Tell Us?9/10/2002 CS 461, Copyright G.D. HagerIllusions: What Do They Tell Us?9/10/2002 CS 461, Copyright G.D. HagerProblems of Computer Vision: Stereo VisionRIGHT IMAGEPLANELEFT IMAGEPLANERIGHTFOCALPOINTLEFTFOCALPOINTBASELINEdFOCALLENGTHfFrom two (or more) images,determine the geometry ofthe scene by matchingcorresponding areas ofthe images9/10/2002 CS 461, Copyright G.D. HagerRandom Dot StereoGram9/10/2002 CS 461, Copyright G.D. HagerTHE ORGANIZATION OF AN IMAGE SEQUENCEFramesFrames areacquired at 30Hz(NTSC)Frames are composed oftwo fields consistingof the even and oddrows of a frame9/10/2002 CS 461, Copyright G.D. HagerTHE MOTION FIELDThe “instantaneous” velocity of points in an imageLOOMINGThe focus of expansionWith just this informationit is possible to calculate:1. Direction of motion2. Time to collision9/10/2002 CS 461, Copyright G.D. HagerMOVING CAMERAS ARE LIKE STEREOLocations ofpoints on the object(the “structure”)The change in spatial locationbetween the two cameras (the “motion”)9/10/2002 CS 461, Copyright G.D. HagerTHE EPIPOLAR CONSTRAINTxan observed pointline along whichthe physical pointmust lie (projection line)the image of theprojection lineXx9/10/2002 CS 461, Copyright G.D. HagerAn Example (Courtesy Carlo Tomasi)9/10/2002 CS 461, Copyright G.D. HagerProblems of Computer Vision: RecognitionGiven a database ofobjects and an imagedetermine what, if anyof the objects are present in the image.9/10/2002 CS 461, Copyright G.D. HagerProblems of Computer Vision: RecognitionGiven a database ofobjects and an imagedetermine what, if anyof the objects are present in the image.9/10/2002 CS 461, Copyright G.D. HagerApplications of Computer Vision: Biometrics• Face recognition• Iris scanning• Fingerprint recognition• Activity recognition9/10/2002 CS 461, Copyright G.D. HagerApplications of Computer Vision: Medical Imaging9/10/2002 CS 461, Copyright G.D. HagerApplications of Computer Vision: Medical ImagingEndoscopic