Development of an automated Moon observation system using the ALTS-07 Robotic Telescope: 2. Progress report on standard contrast enhancement of Moon crescent image with OpenCV
Open Access
- 1 February 2022
- conference paper
- Published by IOP Publishing in Journal of Physics: Conference Series
- Vol. 2214 (1) , 012004
Abstract
Vision is a continuous two-dimensional (2-D) image that can be perceived by human visual system. Mathematically, image is a 2-D function that expresses the intensity of light. This research introduces the fundamental ideas of computer vision. We used OpenCV (Open-Source Computer Vision) to solve automatic image processing techniques, especially enhancement of contrast crescent visibility. We described an image using feature vectors to characterize and numerically to quantify the contents of an image. Then, we compared several techniques for increasing image quality, such as Basic contrast, Binary thresholding, Otsu thresholding, Histogram equalization, and Adaptive Histogram equalization. Finally, we calculated the statistical features of the image before and after enhancement process, and also decided to choose the best enhancement technique that will be implemented on a self-constructed cascade classifier in the future.Keywords
This publication has 7 references indexed in Scilit:
- Analysis of the Quantization Noise in Discrete Wavelet Transform Filters for 3D Medical ImagingApplied Sciences, 2020
- Galaxy detection and identification using deep learning and data augmentationAstronomy and Computing, 2018
- Rotation-invariant convolutional neural networks for galaxy morphology predictionMonthly Notices of the Royal Astronomical Society, 2015
- Galaxy Zoo 2: detailed morphological classifications for 304 122 galaxies from the Sloan Digital Sky SurveyMonthly Notices of the Royal Astronomical Society, 2013
- Real-time computer vision with OpenCVCommunications of the ACM, 2012
- Method of Passive Image Based Crater Autonomous DetectionChinese Journal of Aeronautics, 2009
- Image quality assessment: from error visibility to structural similarityIEEE Transactions on Image Processing, 2004