This paper presents a novel approach for a position verification system in medical applications. By replacing the already existing cross line laser projectors with galvo- or MEMS-based projectors and utilizing the surveillance cameras, a self-calibration of the system is performed and surface acquisition for positioning verification is demonstrated. The functionality is shown by analyzing the radii of calibration spheres and determining the quality of the captured surface with respect to a reference model. The paper focuses on the demonstration with one pair of camera and projector but can also be extended to a multi-camera-projector system, as present in treatment rooms. Compared to other systems, this approach does not need external hardware and is thus space and cost efficient.

In recent years, smartphones have become the primary device for day-to-day photography. Therefore, it is critical for mobile imaging to capture sharp images automatically without human intervention. In this paper, we formulate autofocus as a decisionmaking process, in which the travel distance of a lens is determined from the phase data obtained from the phase sensors of a smartphone, and the decision-making policy is based on reinforcement learning, a popular technique in the field of deep learning. We propose to use a noise-tolerant reward function to combat the noise of the phase data. In addition, instead of using only the current phase data, each lens movement is determined using the phase data acquired along the journey of an autofocus process. As a result, the proposed machine-learning approach is able to expedite the autofocus process as well. Experimental results show that the method indeed improves the autofocus speed.

In face-priority automatic exposure control in digital camera systems, exposure adjustment is typically made irrespective of the face skin tone, i.e. how dark or bright the face is. As a result, a face can become over exposed when it is present against a very dark background, or become under exposed when it is present against a very bright background, depending on the face skin tone and scene content. Adapting the exposure control to the face skin tone will result in well-exposed faces in various capture scenarios, and hence better image quality.This paper presents a novel face skin tone adaptive automatic exposure control solution. Using a well-trained neural network based face skin tone predictor, the likelihoods of dark and bright face skin tones are calculated. An algorithm adjusts the bounds of the target brightness of the exposure control based on the face skin tone likelihoods and a set of configuration parameters. The face skin tone adaptive brightness bounds then guide the frame exposure adjustment. Experimental results demonstrate the outperformance of the proposed solution over conventional exposure control that does not take into account the face skin tone information.

Stack-based high dynamic range (HDR) imaging is a technique for achieving a larger dynamic range in an image by combining several low dynamic range images acquired at different exposures. Minimizing the set of images to combine, while ensuring that the resulting HDR image fully captures the scene’s irradiance, is important to avoid long image acquisition and postprocessing times. The problem of selecting the set of images has received much attention. However, existing methods either are not fully automatic, can be slow, or can fail to fully capture more challenging scenes. In this paper, we propose a fully automatic method for selecting the set of exposures to acquire that is both fast and more accurate. We show on an extensive set of benchmark scenes that our proposed method leads to improved HDR images as measured against ground truth using the mean squared error, a pixel-based metric, and a visible difference predictor and a quality score, both perception-based metrics.

Since the introduction of the Minolta Maxxum 9000 in 1985, PDAF (phase detect automatic focus) has been the standard way to achieve sharply-focused images of fast-moving action, such as professional sports. In a typical SLR (single lens reflex) camera, the image for the optical viewfinder is reflected up by the main mirror, while a secondary mirror and optics copy the image to the PDAF detector. However, such an arrangement is impractical for mirrorless digital cameras. Thus, there have been a variety of methods used to incorporate phase sensing on the main sensor – with various trade-offs. The current work discusses some of these trade-offs and then describes in detail a specific type of striping artifact introduced by the masked pixel structures used in Sony sensors. A computational method for credible repair of this artifact also is presented. The method described is quick and fully automatic; it has been implemented as KARWY-SR, an open source JavaScript version using a drag-and-drop interface to repair the artifact in Sony ARW raw files.

This work was carried out to serve two purposes:• Create and share a single motion profile that emulates the handshake of a population of mobile phone users taking still photos under real life conditions.• Describe the validation procedure required to ensure the high fidelity motion platform chosen can correctly reproduce the proposed motion profile.By means of psychophysical testing, we examined the frequency and spatial characteristics of human handshake, based on which we created synthetic handshake profile with very similar properties.We demonstrate how the proposed motion trace correlates extremely well with real handshake and why using a realistic motion profile together with a high fidelity motion platform is crucially important in order to avoid disturbances not present with real users.

The mobile and Smartphone camera market is driven by consumers' desire for better picture quality. Auto-flash is the most common flash mode setting used by consumers on their smartphones. Auto-flash technique is used in modern cameras to turn on the flash light automatically when the light level is lower than a pre-defined level. The auto-flash activation light level varies among different cameras but should be set to a value that provides a balance between good picture quality and less battery consumption. There is no published methodology on how to determine the correct light level for auto-flash activation, so I developed one based on Signal to Noise Ratio analysis, SNR10, and image sharpness analysis that determines the light settings for autoflash mode. Then the aforementioned method was tested using the Apple iPhone and Samsung Galaxy. Finally, I determined the auto-flash light activation level for these phones using this methodology.

Improving image quality by capturing high dynamic range (HDR) in a scene remains a technological challenge for CMOS sensors. The multi partial reset is a simple technique that allows a sensor to capture HDR inexpensively with high frame rate, low noise floor, and high signal to noise ratio. However, it flattens images due to compression of highlights, rending it useful only for machine vision and automotive applications. In this work, we present an inverse function that restores HDR image appearance specifically for a multi partial reset sensor (MPRS). This function can be applied in software or firmware before demosaicing. Results show that the function automatically enliven images with more depth and saturation that suit general purpose photography. Moreover, latency results show that it can be applied for real-time videography of high frame rate. These results would be computationally much more expensive to achieve using general image enhancement techniques, i.e. not specific to MPRS, especially for high definition, high frame rate, real-time video.

When exposure times were measured in minutes, the opening and closing of the shutter was essentially instantaneous. As more sensitive films and brighter optics became available, exposure times decreased, the travel time of the shutter mechanism became increasingly significant, and artifacts became visible. Perhaps the best-known shutter artifacts are the spatio-temporal distortions associated with photographing moving subjects using a focal-plane shutter or sequential electronic sampling of pixels (electronic rolling shutter). However, the shutter mechanism also can cause banding with flickering light sources and strange artifacts in out-of-focus regions (bokeh); it can even impact resolution. This paper experimentally evaluates and discusses the artifacts caused by leaf, focal plane, electronic first curtain, and fully electronic sequential-readout shuttering.