Article Details

Application of image contours to three aspects of image processing : compression, shape recognition and stereopsis

Oleh   S. Marshall, BSc, PhD, CEng, MlEE [-]
Kontributor / Dosen Pembimbing : -
Jenis Koleksi : Prosiding
Penerbit : Lain-lain
Fakultas :
Subjek :
Kata Kunci : Image processing, Picture processing and pattern recognition
Sumber : IEE PROCEEDINGS-I, Vol. 139, No. I , FEBRUARY 1992
Staf Input/Edit : Irwan Sofiyan  
File : 1 file
Tanggal Input : 2019-01-21 09:02:38

Image processing comprises two distinct areas of research, namely image analysis and image compression. However, the thrust of research towards the analysis of digital images and towards their efficient coding has been pursued almost independently. Both areas are of vital importance. The prospect of building ‘machines that see’ for industrial inspection, medical imaging etc. is an exciting one indeed. Also the necessity to develop efficient coding techniques to reduce the number of data required for image storage and transmission is clear. In both image analysis and image compression it is necessary to derive a representation from the pixel data. Although basic filtering and template matching approaches in image analysis operate directly upon the image intensity values, more complex tasks such as unconstrained recognition must operate on higher-level primitive descriptions. Exactly what these primitives should comprise is still an area of current research. Features such as edges, corners and zero crossings of the second derivative have been proposed Marr [2]. Other