A defocus blur identification technique based on histogram analysis of a real edge image is presented. The image defocus process of a camera is formulated by incorporating the nonlinear camera response and the intensity-dependent noise model. The histogram matching between the synthesized and real defocused regions is then carried out with intensity-dependent filtering. By iteratively changing the point-spread function parameters, the best blur extent is identified from histogram comparison. We have performed the experiments on both the synthetic and real edge images. The results have demonstrated the robustness and feasibility of the proposed technique.
Download full-text PDF |
Source |
|---|---|
| http://dx.doi.org/10.1364/JOSAA.29.001694 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Fringe projection profilometry (FPP) is a widely adopted technique for three-dimensional (3D) reconstruction. However, its depth-of-field (DOF) is constrained when reconstructing defocused scenes, mainly due to limitations in the camera model and image blur. This study introduces a camera model based on the ideal optical system, which effectively reduces the systematic errors associated with the conventional pinhole camera model.
View Article and Find Full Text PDFOPTICAL BLUR AFFECTS DIFFERENTLY ON AND OFF VISUAL PATHWAYS.
bioRxiv
October 2024
Department of Biological and Visual Sciences, SUNY Optometry, New York, NY 10036, USA.
The human eye has a crystalline lens that focuses retinal images at the point of fixation. Outside this fixation region, images are distorted by optical blur, which increases light scatter and reduces the spatial resolution and contrast processed by neuronal pathways. The spectacle lenses that humans use for optical correction also minify or magnify the images, affecting neuronal surround suppression in visual processing.
View Article and Find Full Text PDFDiffusion Optics Technology (DOT): A Myopia Control Spectacle Lens Based on Contrast Theory.
Transl Vis Sci Technol
October 2024
Department of Ophthalmology, University of Washington, Seattle, WA, USA.
Diffusion optics Technology (DOT) myopia control spectacle lenses are based on contrast theory. This innovative theory represents a radical departure from the classical concept of visual deprivation myopia. However, traditional theories have evolved, arriving at remarkably similar solutions for myopia control as the DOT lenses.
View Article and Find Full Text PDFMotion onset VEPs can see through the blur.
Sci Rep
September 2024
Department of Medical Biophysics, Faculty of Medicine in Hradec Kralove, Charles University, Simkova 870, 500 03, Hradec Kralove, Czechia.
Motion-onset visual evoked potentials (MO VEPs) are robust to dioptric blur when low contrast and low spatial frequency patterns are used for stimulation. To reveal mechanisms of MO VEPs robustness, we studied whether the resistance to defocus persists even when using a high-contrast checkerboard using digital defocus in the emmetropic eyes of 13 subjects (males 20-60 years). We compared the dominant components of MO VEPs to pattern-reversal VEPs (PR VEP), which are sensitive to the blur.
View Article and Find Full Text PDFWeakly-Supervised Depth Estimation and Image Deblurring via Dual-Pixel Sensors.
IEEE Trans Pattern Anal Mach Intell
December 2024
Article Synopsis
- Dual-pixel (DP) imaging sensors are becoming popular in modern cameras, capturing two images per pixel which helps in depth estimation, but defocus blur can hinder accuracy.
- The authors propose a joint approach that combines blur removal and depth estimation using a new model called WDDNet, which operates without the need for ground-truth depth maps.
- Experiments show that WDDNet performs competitively against established supervised methods by using all-in-focus images as supervisory signals for depth estimation and image restoration.
Want AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!