An autofocus heuristic for digital cameras based on supervised machine learning

Mir, Hashim; Xu, Peter; Chen, Rudi; Beek, Peter van der

doi:10.1007/s10732-015-9291-4

Cited by 16 publications

(10 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Specifically, by integrating phase detection AF and contrast AF, as described above, to compensate for the respective AF methods' shortcomings, the speed and accuracy of AF in CSC were improved. The phase detection AF method's drawback is that the actual focal position does not match due to AF computation movement [10]. Therefore, if it is supplemented with contrast AF, that focuses accurately, it is possible to focus on fast-moving objects even in dark surroundings.…”

Section: Introductionmentioning

confidence: 99%

Design of Wide Angle and Large Aperture Optical System with Inner Focus for Compact System Camera Applications

Choi

Ryu

2019

Applied Sciences

View full text Add to dashboard Cite

Conventionally, a bright, very wide-angle optical system is designed as a floating type optical system that moves two or more lens groups composed of multiple lens in order to focus accurately. These have been widely used as phase detection auto focus (AF) methods within conventional digital single-lens reflex (DSLR) cameras. However, a phase detection AF optical system cannot be used when recording motion pictures. In contrast, a compact system camera (CSC) performs AF by the contrast method, where a stepper motor is used as the driving source for moving the optical lens. Nonetheless, to ensure that the focusing lens is lighter, these stepper motors should not have high torque and AF must be possible by moving only one lens. Yet, when focusing is performed with only one lens, aberration change due to focusing lens movement is magnified. Therefore, a very wide-angle optical system comprised of a half-angle of view more than 40 degrees and F of 1/4 has not been developed. Here, a very wide-angle optical system was designed with high resolving power that enables high speed AF, even in contrast mode, by moving only one lens while minimizing aberration change.

show abstract

Section: Introductionmentioning

confidence: 99%

Design of Wide Angle and Large Aperture Optical System with Inner Focus for Compact System Camera Applications

Choi

Ryu

2019

Applied Sciences

View full text Add to dashboard Cite

show abstract

“…Mir et al . (), unlike above publications, considered the case of multimodal model and implemented a machine learning‐based coarse‐to‐fine search to find all the peaks, typically the foreground and background. This method does not fall with the scope of this study, which is the unimodality of the model; indeed, our specimens are such they can be imaged at one time by the SEM.…”

Section: Introductionmentioning

confidence: 99%

“…It is easier to implement than the method of Mir et al . (), since it is just a regression and only requires an off‐line automatic training to determine its parameters. Finally, the solution is an efficient autofocus method that works well at low and high magnifications.…”

Section: Introductionmentioning

confidence: 99%

Combining gradient ascent search and support vector machines for effective autofocus of a field emission–scanning electron microscope

Dembélé

Lehmann

Medjaher

et al. 2016

Journal of Microscopy

View full text Add to dashboard Cite

Autofocus is an important issue in electron microscopy, particularly at high magnification. It consists in searching for sharp image of a specimen, that is corresponding to the peak of focus. The paper presents a machine learning solution to this issue. From seven focus measures, support vector machines fitting is used to compute the peak with an initial guess obtained from a gradient ascent search, that is search in the direction of higher gradient of focus. The solution is implemented on a Carl Zeiss Auriga FE-SEM with a three benchmark specimen and magnification ranging from x300 to x160 000. Based on regularized nonlinear least squares optimization, the solution overtakes the literature nonregularized search and Fibonacci search methods: accuracy improvement ranges from 1.25 to 8 times, fidelity improvement ranges from 1.6 to 28 times, and speed improvement ranges from 1.5 to 4 times. Moreover, the solution is practical by requiring only an off-line easy automatic train with cross-validation of the support vector machines.

show abstract

“…Other types of strategies are motivated by reducing the time and stack size to find the image with the best focus (e.g., maximizing the number of ROIs in focus). These strategies are based on optimization methods to estimate some extreme point of a function (e.g., maximum of a focus function) [32,35,36,57]. • Stitching: This is the process carried out to obtain an image composed by several images with overlapping FOVs.…”

mentioning

confidence: 99%

“…• (1-DIG) Multifocus and multiexposure fusion: When several images are captured at different focal planes, the inherent three-dimensional morphology of diatom valves is recorded and could be fused in a new complete in focus image. Similarly, multiexposure fusion is used to obtain the right exposure over all regions in the fused image from individual images with underexposed and overexposed regions[9,10,13,[30][31][32][33][34][35][36].• …”

mentioning

confidence: 99%

A Low-Cost Automated Digital Microscopy Platform for Automatic Identification of Diatoms

et al. 2020

View full text Add to dashboard Cite

Currently, microalgae (i.e., diatoms) constitute a generally accepted bioindicator of water quality and therefore provide an index of the status of biological ecosystems. Diatom detection for specimen counting and sample classification are two difficult time-consuming tasks for the few existing expert diatomists. To mitigate this challenge, in this work, we propose a fully operative low-cost automated microscope, integrating algorithms for: (1) stage and focus control, (2) image acquisition (slide scanning, stitching, contrast enhancement), and (3) diatom detection and a prospective specimen classification (among 80 taxa). Deep learning algorithms have been applied to overcome the difficult selection of image descriptors imposed by classical machine learning strategies. With respect to the mentioned strategies, the best results were obtained by deep neural networks with a maximum precision of 86% (with the YOLO network) for detection and 99.51% for classification, among 80 different species (with the AlexNet network). All the developed operational modules are integrated and controlled by the user from the developed graphical user interface running in the main controller. With the developed operative platform, it is noteworthy that this work provides a quite useful toolbox for phycologists in their daily challenging tasks to identify and classify diatoms.

show abstract

An autofocus heuristic for digital cameras based on supervised machine learning

Cited by 16 publications

References 21 publications

Design of Wide Angle and Large Aperture Optical System with Inner Focus for Compact System Camera Applications

Design of Wide Angle and Large Aperture Optical System with Inner Focus for Compact System Camera Applications

Combining gradient ascent search and support vector machines for effective autofocus of a field emission–scanning electron microscope

A Low-Cost Automated Digital Microscopy Platform for Automatic Identification of Diatoms

Contact Info

Product

Resources

About