Neuroevolutionary digital image processing

V. Spitsyn, Y. Tsoy

Результат исследований: Материалы для книги/типы отчетовМатериалы для конференции

Аннотация

An approach to image enhancement through artificial neural network's (ANN) processing is proposed. The structure and weights of ANN are tuned with use of evolutionary concept. Each image is processed in pixel-by-pixel manner using pixels' local characteristics that are calculated approximately to increase the processing speed but preserving satisfactory calculations' error. The three-step procedure for image enhancement is proposed: (1) image brightness preprocessing; (2) local level processing using ANN; (3) global level autoleveling algorithm. The results for the proposed three-step image enhancement procedure are presented.

Язык оригиналаАнглийский
Название основной публикацииProceedings of the 2009 International Conference on Image Processing, Computer Vision, and Pattern Recognition, IPCV 2009
Страницы833-838
Число страниц6
Том2
СостояниеОпубликовано - 2009
Событие2009 International Conference on Image Processing, Computer Vision, and Pattern Recognition, IPCV 2009 - Las Vegas, NV, Соединенные Штаты Америки
Продолжительность: 13 июл 200916 июл 2009

Другое

Другое2009 International Conference on Image Processing, Computer Vision, and Pattern Recognition, IPCV 2009
СтранаСоединенные Штаты Америки
ГородLas Vegas, NV
Период13.7.0916.7.09

ASJC Scopus subject areas

  • Computer Graphics and Computer-Aided Design
  • Computer Vision and Pattern Recognition

Fingerprint Подробные сведения о темах исследования «Neuroevolutionary digital image processing». Вместе они формируют уникальный семантический отпечаток (fingerprint).

Цитировать