Performance Of Autofocus Research Articles

Enhancing Autofocus in Non-Mydriatic Fundus Photography: A Fast and Robust Approach with Adaptive Window and Path-Optimized Search

Non-mydriatic fundus photography (NMFP) plays a vital role in diagnosing eye diseases, with its performance primarily dependent on the autofocus process. However, even minor maloperations or eye micro-movements can compromise fundus imaging quality, leading to autofocus inaccuracy and a heightened risk of misdiagnosis. To enhance the autofocus performance in NMFP, a fast and robust fundus autofocus method with adaptive window and path-optimized search is proposed. In this method, the adaptive focus window is used to suppress irrelevant image contents and correct the sharpness curve, and the path-optimized search is constructed to overcome the curve’s local extrema, in order to achieve rapid focus position convergence. This method was simulated and clinically studied with the self-developed autofocus system for NMFP. The results of 80 cases of human eye imaging show that, compared with similar autofocus methods, this method achieves a focus success rate of 90% with the least axial scanning, and can adapt to non-ideal imaging conditions such as pupil misalignment, eyelash occlusion, and nystagmus.

Abrupt autofocusing performance of a circular Airyprime beam with vortex pairs

Abrupt autofocusing performance of a circular Airyprime beam (CAPB) with vortex pairs propagating in free space is investigated. Here the vortex pairs are off-axis by default. The influences of topological charges and spatial location of vortex pairs on abrupt autofocusing behaviour of a CAPB with vortex pairs are evaluated by numerical simulations. A intensity contrast, which is the ratio of peak intensity on a certain observation plane to that on initial plane, is used to describe the abrupt autofocusing ability. The focus length of a CAPB with vortex pairs only depends on topological charges and is independent of spatial location of vortex pairs. When the topological charges increase, the abrupt autofocusing ability is enhanced, which is accompanied by the slight extension of the focal length. The spatial location of vortex pairs includes azimuth and radial locations. The effect of the azimuth location of vortex pairs on the abrupt autofocusing behaviour is so slight that it can be neglected. If the radial location of vortex pairs decreases, the abrupt autofocusing ability is improved. Abrupt autofocusing performance of a CAPB with vortex pairs is also compared with that of a CAPB with an optical vortex. Due to large dark-hollow intensity distribution and carrier of much orbital angular momentum, a CAPB with vortex pairs has advantages over a CAPB with an off-axis optical vortex in optical capture and other applications.

A method for evaluating camera auto-focusing performance using a transparent display device

With the development of various autofocusing (AF) technologies, sensor manufacturers are demanded to evaluate their performance accurately. The basic method of evaluating AF performance is to measure the time to get the refocused image and the sharpness of the image while repeatedly inducing the refocusing process. Traditionally, this process was conducted manually by covering and uncovering an object or sensor repeatedly, which can lead to unreliable results due to the human error and light blocking method. To deal with this problem, we propose a new device and solutions using a transparent display. Our method can provide more reliable results than the existing method by modulating the opacity, pattern, and repetition cycle of the target on the transparent display.

SAR Image Reconstruction and Autofocus Using Complex-Valued Feature Prior and Deep Network Implementation

Synthetic aperture radar (SAR) plays an important role in remote sensing by providing electromagnetic images of the observation scene. The prior knowledge-based SAR image reconstruction method can reduce the requirement of data sampling ratio and improve image quality. The existing prior knowledge-based method usually uses the magnitude information of SAR images while ignoring the phase information. However, since the echo and backscattering coefficients are complex values, the phase information of SAR images will help improve the reconstruction accuracy in the image reconstruction process. To improve the reconstruction performance, this paper proposes a SAR image reconstruction and autofocus method using complex-valued feature prior. In the proposed method, the complex-valued feature prior is learned from data by a complex-valued feature projection operator (CFPO), which can characterize and extract scene features in Range-Doppler domain and 2D frequency domain. The proposed CFPO enables more efficient use of echo data and helps to improve image reconstruction and autofocus performance. In addition, the proposed method is implemented by an unfolded deep network, which enables data-driven feature learning and efficient computation. The proposed method is verified by simulated and measured data.

Quantitative Evaluation Method for Image Autofocusing Performance

Quantitative Evaluation Method for Image Autofocusing Performance

Image Reconstruction Using Autofocus in Single-Lens System

To reconstruct the wavefront in a single-lens coherent diffraction imaging (CDI) system, we propose a closed-loop cascaded iterative engine (CIE) algorithm based on the known information of the imaging planes. The precision of diffraction distance is an important prerequisite for a perfect reconstruction of samples. For coherent diffraction imaging with a lens, autofocus is investigated to accurately determine the object distance and image distance. For the case of only the object distance being unknown, a diffuser is used to scatter the coherent beam for speckle illumination to improve the performance of autofocus. The optimal object distance is obtained stably and robustly by combing speckle imaging with clarity evaluation functions. SSIM and MSE, using the average pixel value of the reconstructed data set as a reference, are applied on two-unknown-distance autofocus. Simulation and experiment results are presented to prove the feasibility of the CIE and proposed auto-focusing method.

Robust autofocusing for scanning electron microscopy based on a dual deep learning network

Scanning electron microscopy (SEM) is a high-resolution imaging technique with subnanometer spatial resolution that is widely used in materials science, basic science, and nanofabrication. However, conducting SEM is rather complex due to the nature of using an electron beam and the many parameters that must be adjusted to acquire high-quality images. Only trained operators can use SEM equipment properly, meaning that the use of SEM is restricted. To broaden the usability of SEM, we propose an autofocus method for a SEM system based on a dual deep learning network, which consists of an autofocusing-evaluation network (AENet) and an autofocusing-control network (ACNet). The AENet was designed to evaluate the quality of given images, with scores ranging from 0 to 9 regardless of the magnification. The ACNet can delicately control the focus of SEM online based on the AENet’s outputs for any lateral sample position and magnification. The results of these dual networks showed successful autofocus performance on three trained samples. Moreover, the robustness of the proposed method was demonstrated by autofocusing on unseen samples. We expect that our autofocusing system will not only contribute to expanding the versatility of SEM but will also be applicable to various microscopes.

Microscope autofocus algorithm based on number of image slope variations.

This paper presents a passive autofocus algorithm applicable to interferometric microscopes. The proposed algorithm uses the number of slope variations in an image mask to locate the focal plane (based on focus-inflection points) and identify the two neighboring planes at which fringes respectively appear and disappear. In experiments involving a Mirau objective lens, the proposed algorithm matched the autofocusing performance of conventional algorithms, and significantly outperformed detection schemes based on zero-order interference fringe in dealing with all kinds of surface blemish, regardless of severity. In experiments, the proposed algorithm also proved highly effective in cases without fringes.

AF-Net: A Convolutional Neural Network Approach to Phase Detection Autofocus

It is important for an autofocus system to accurately and quickly find the in-focus lens position so that sharp images can be captured without human intervention. Phase detectors have been embedded in image sensors to improve the performance of autofocus; however, the phase shift estimation between the left and right phase images is sensitive to noise. In this paper, we propose a robust model based on convolutional neural network to address this issue. Our model includes four convolutional layers to extract feature maps from the phase images and a fully-connected network to determine the lens movement. The final lens position error of our model is five times smaller than that of a state-of-the-art statistical PDAF method. Furthermore, our model works consistently well for all initial lens positions. All these results verify the robustness of our model.

Compressive Sensing Based SAR Imaging and Autofocus Using Improved Tikhonov Regularization

For synthetic aperture radar (SAR) imaging, the compressive sensing (CS) coupled with total variation (TV)-based algorithm is known as an effective focusing technique using under-sampled dataset. However, the performance of the CS-TV method can be degraded by drawbacks of TV in terms of noise sensitivity and computational efficiency. In this paper, a novel approach to CS-SAR imaging is proposed based on improved Tikhonov regularization (ITR) coupled with an adaptive strategy using iterative reweighted matrix to solve the CS reconstruction problem of SAR images with sparsity. The proposed method can provide different degrees of performance of SAR autofocus with changes to the value of certain parameters of ITR. The proposed scheme outperforms conventional CS-based methods with respect to image quality, noise robustness, and computational complexity of the algorithm owing to the additional sensitivity of the proposed objective function. From the simulation results, we verify that the proposed autofocus method is highly efficient in forming SAR images from non-uniformly under-sampled dataset in terms of both image quality and computational efficiency.

Finding autofocus region in low contrast surveillance images using CNN-based saliency algorithm

How to automatically locate the focus region in low contrast image is a key issue for camera-equipped surveillance devices. Due to low signal to noise ratio, the performance of autofocus will seriously decline in low contrast image, making it quite difficult to recognize the focus region. To tackle this problem, we perform autofocus by conducting a salient object detection approach. A covariance based deep learning framework is proposed to evaluate the saliency of low contrast surveillance image. Based on the mechanism of human visual system, the autofocus region can be identified by the visual salient object. In this paper, low-level features of the low contrast images are first studied and extracted. Then the mutual covariances of the segmented blocks are trained via a 7-layers convolutional neural network (CNN). Next, the initial saliency map of the testing image can be obtained by estimating the saliency score of each block via the pre-trained CNN model. Finally, the resulting saliency map is refined by introducing the local-global difference and internal similarity approaches. Experimental results demonstrate that the proposed method outperforms existing ten state-of-the-art saliency models on three public datasets and a nighttime image dataset.

Performance of differential synthetic aperture ladar versus phase-gradient autofocus at large phase errors and low return levels

The performance of differential synthetic aperture ladar (DSAL) processing versus phase-gradient autofocus (PGA) technique in SAL image formation is demonstrated experimentally using phase history data (PHD) with large random phase errors and various target returns generated from a laboratory DSAL setup operating at 1550-nm wavelength. Results show that at strong return levels, both DSAL and PGA are very robust in focusing high-resolution images from the PHD; whereas, as target returns become weak and PHD is dominated by noise, the iterative PGA performs less effectively than the DSAL does. It is concluded that, on the whole, the DSAL technique is possibly more robust than PGA in both effectiveness and efficiencies.

Simplified Numerical Model for Rapid Design of AF Module Spring Shape Parameters

AbstractThe autofocusing (AF) performance of cell phone cameras is critically dependent on the design of the voice-coil motor (VCM) used to drive the lens module. Also, the metal springs in the AF module should combine high stiffness with a good actuation response and a light weight. The present study utilizes a reverse engineering approach to construct three-dimensional finite element models of the top and bottom springs in the VCM mechanism. Simulations are then performed to investigate the von Mises stress distribution and stiffness characteristics of the two springs given horizontal and vertical orientations of the AF module, respectively. In performing the simulations, the actuation force is computed using two different analysis methods, namely a simplify structure method and a coupled electromagnetic-structural method. It is shown that the simplify structure method has the advantages of a lower computational complexity and a more comprehensive modeling capability. A further series of simulations is thus to examine the effects of the spring shape parameters on the reaction force developed by the spring stiffness. The results show that the spring stiffness increases with an increasing thickness and a decreasing rib length. The simulation results obtained for different spring shape parameter settings are summarized in the form of a parameter design chart for predicting the reaction force given known values of the spring rib length and spring thickness.

Auto Focus Performance –What can we expect from todays cameras?

Auto Focus Performance –What can we expect from todays cameras?

Autofocus measurement for imaging devices

We propose an objective measurement protocol to evaluate the autofocus performance of a digital still camera. As most pictures today are taken with smartphones, we have designed the first implementation of this protocol for devices with touchscreen trigger. The lab evaluation must match with the users' real-world experience. Users expect to have an autofocus that is both accurate and fast, so that every picture from their smartphone is sharp and captured precisely when they press the shutter button. There is a strong need for an objective measurement to help users choose the best device for their usage and to help camera manufacturers quantify their performance and benchmark different technologies.

BEMD-based two-dimensional SAR autofocus algorithm

The two-dimensional (2D) autofocus algorithm simultaneously compensates both the residual range migration and propagation-induced phase errors. However, the unsuitable selection of prominent points would reduce the performance of autofocus. Bi-dimensional empirical mode decomposition (BEMD) can decompose a synthetic aperture radar (SAR) image into some different sub-images corresponding to different frequency components. Using proposed 2D autofocus algorithm along with BEMD would help us in selecting the suitable prominent points and give a considerable improvement in the performance of SAR autofocus. In further support of that conclusion, some simulations using real SAR data are demonstrated.

Region sampling for robust and rapid autofocus in microscope.

This paper proposes a region sampling based autofocus method for rapid and robust autofocus in microscope. Image content and region size are considered in region sampling criteria. An intelligent search algorithm which employs quartering hill climbing search and golden section search is developed, in which rule-based evaluation of sampled focusing regions is applied. Experimental results demonstrate that the proposed method can significantly improve the performance of image-based autofocus.

Improving the accuracy and low-light performance of contrast-based autofocus using supervised machine learning

The passive autofocus mechanism is an essential feature of modern digital cameras and needs to be highly accurate to obtain quality images. In this paper, we address the problem of finding a lens position where the image is in focus. We show that supervised machine learning techniques can be used to construct heuristics for a hill-climbing approach for finding such positions that out-performs previously proposed approaches in accuracy and robustly handles scenes with multiple objects at different focus distances and low-light situations. We gather a suite of 32 benchmarks representative of common photography situations and label them in an automated manner. A decision tree learning algorithm is used to induce heuristics from the data and the heuristics are then integrated into a control algorithm. Our experimental evaluation shows improved accuracy over previous work from 91.5% to 98.5% in regular settings and from 70.3% to 94.0% in low-light.

Noise insensitive focus value operator for digital imaging systems

Most auto-focusing algorithms based on focus values which are calculated with high frequency image components are very sensitive to noise factor. The noise under low illumination conditions generally degrades the auto-focusing performance. In this paper, we propose an operator to determine the noise insensitive focus value for digital imaging systems. The proposed algorithm uses spatially adapted weights based on the local statistics of the image and the noise factors. Experiments conducted with and without noise factors demonstrate the high performance of the proposed method.

Dynamic evaluation of autofocusing for automated microscopic analysis of blood smear and pap smear

Autofocusing is a fundamental procedure towards automated microscopic evaluation of blood smear and pap smear samples for clinical diagnosis. This paper presents comparison results of 16 selected focus algorithms based on 8000 static bright-field images and 1600 dynamic autofocusing trials using 10 blood smear and pap smear samples. Besides static behaviour, dynamic autofocusing performance is introduced for ranking the 16 focus algorithms. The Fibonacci search algorithm is employed for controlling the z-motor of the microscope to reach the focus position that is determined by focus objective functions. Experimental results demonstrate that the variance algorithm provides the best overall performance. Together with our previously reported findings, it is demonstrated that the variance algorithm or the normalized variance algorithm is the optimal focus algorithm for non-fluorescence microscopy applications including pap smear and blood smear imaging.