Search

Many telepresence robots are equipped with a forward-facing camera for video communication and a downward-facing camera for navigation. In this paper, we propose to stitch videos from the FF-camera with a wide-angle lens and the DF-camera with a fisheye lens for telepresence robots. We aim at providing a compact and efficient user interface for telepresence robots with user-friendly interactive experiences. To this end, we present a multi-homography-based video stitching method which stitches videos from a wide-angle camera and a fisheye camera. The method consists of video image alignment, seam cutting, and image blending. We directly align the wide-angle video image and the fisheye video image based on the multi-homography alignment without calibration, distortion correction, and unwarping procedures. Thus, we can obtain a stitched video with shape preservation in the non-overlapping regions and alignment in the overlapping area for telepresence. To alleviate ghosting effects caused by moving objects and/or moving cameras during telepresence robot driving, an optimal seam is found for aligned video composition, and the optimal seam will be updated in subsequent frames, considering spatial and temporal coherence. The final stitched video is created by image blending based on the optimal seam. We conducted a user study to demonstrate the effectiveness of our method and the superiority of user interfaces with a stitched video.

Most computer vision research focuses on narrow angle lenses and is not adapted to super-wide-angle (aka spherical) lenses. This is mainly because current neural networks are not designed or trained to interpret the significant barrel distortion that is introduced in the captured image by such wide angle lenses. As these lenses capture a half-sphere or a section of sphere on the object space, barrel distortion appears when the image is projected on a 2D flat image sensor. By controlling this distortion at the lens design stage, camera designers can create some areas with augmented resolution [26]. In this work, we present an analysis of the impact of such augmented resolution on computer vision algorithm accuracy, using the problem of single image depth estimation as a case study. To this end, 360° panorama datasets are warped to simulate different wide-angle lens datasets, which are then used to train identical neural networks. Each lens presents specific areas of the image with augmented resolution using spatially-varying non-linear distortion. We show that this property leads to better local accuracy in depth estimation. We also demonstrate that considering lens manufacturing improves performance when tested on realistic lenses, especially in the area of augmented resolution. We further show that this property helps to locally come closer to performances obtained on perspective images without cropping the field of view.

Wide-angle lenses typically offer fields of view greater than 70°, which are utilized in a variety of imaging, mapping, and navigation applications. Wide-angle lenses are commonly modelled using the central perspective model, compensating for lens distortions through a series of additional parameters. The more extreme the distortions, the further the reality of the lens matches the collinearity equations that define the central perspective model. Fisheye lenses are modelled differently because their fields of view are so wide (typically 180°) that the collinearity model is not applicable. This work studied the effects of modelling wide-angle lenses using both the conventional central perspective model and the fisheye model to determine which model best fits the observations and models the distortions more precisely and accurately. These results were produced by generating observations in a dedicated indoor calibration facility at the University of Calgary: an 11 m × 11 m × 4 m field comprising 291 signalized photogrammetry targets. Multiple free-network, self-calibrating bundle adjustments were performed using different models and different cameras. The results of the self-calibrating bundle adjustments were then utilized in a check adjustment on independent sets of check images to validate their accuracy. Two cameras, a Ladybug5 and a GoPro Hero5, were tested. The GoPro was also calibrated using a checkerboard target pattern, and the results were compared to those of the 3D calibration target-field. The results of the bundle adjustments determined that the fisheye model describes the distortions more precisely in both wide-angle camera systems.

As camera technology develops, products have improved and people have an increasing demand for visual quality. To increase the field of view, many zooming modules have been designed, which has raised the cost of optic components, increased the amount of calculation required, and complicated image processing technology. To date, dashcam lenses are primarily a compensatory design that cannot autozoom. Developing technology with various focuses for use on different roads has become a critical topic for autopilot development. Possibilities include using a short focus on general roads and using a long focus for freeways. Dashcam lens design requires the ability to cover wide angles. The concept behind wide-angle zooming lenses is that when the focal length is the shortest, the widest monitoring range is achieved. To observe a point at a far distance, we can use optical zoom to increase the focal length, thereby enabling a view into the distance. At this time, the angular field of view is sacrificed. However, a reduction of the angular field of view leads to a reduction of distortion, which subsequently reduces the complexity of back-end image processing and leads to clearer images. The design in this study references the wide-angle lens patent US.20090080093 (Ning in Compact fisheye objective lens, U.S., Patent No. 20090080093, 2011) of Alex Ning, whose optical system consists of six spherical lenses. The angular field of view reached 170°. The total system length is 18 mm, and F/# is 3. The system focus is 1.652 mm. Four groups of liquid lenses were placed in this wide-angle lens. Altering the curvature and thickness of the liquid lenses changes the course of light and forms a new angular field of view. This system consists of three zoom settings, with fields of view being 170°, 160°, and 150°, respectively. The F/# was 2.4. The modulation transfer function at a spatial frequency of 1801 p/mm reached 6%, 0%, and 27%, respectively. F-Theta was controlled within ± 10%. The spot size was smaller than 4 μm, and the field curvature was smaller than ± 0.02 mm. The liquid lenses were successfully introduced into the wide-angle lens to achieve optical zoom without using a conventional mobile mechanical structure.

In the lead article of the May-June 2021 issue of the Hastings Center Report, Nancy Jecker and Caesar Atuire argue that the Covid-19 crisis is best understood as a syndemic, "a convergence of biosocial forces that interact with one another to produce and exacerbate clinical disease and prognosis." A syndemic framework, the authors advise, will enable bioethicists to recognize the ethical principles that should guide efforts to reduce the unequal effects that Covid-19 has on populations. Drawing on sub-Saharan African conceptions of solidarity, the authors lay out an approach to global vaccine distribution that prioritizes low- and middle-income countries. Like Jecker and Atuire's article, an essay by philosopher Keisha Ray pushes bioethicists to recognize broader justice-oriented responsibilities with the aid of a wide-angle lens. Ray's essay focuses on contemporary examples of environmental injustices that sicken, disable, or kill Black people.

In today’s world, increasing number of terror and intruders at the line of control becoming a huge threat for the country and it is not possible for a solider to watch every moment of time that paves a way for the terror activities in country, in order to avoid such activities more number of soldiers are involved in the border due to which many soldiers’ lives are at risk. A robot named Smart Delimit Scrutinization Droid (SD2) has been developed as a solution for the above issue, it is a robot which takes the role of soldiers for the border surveillance, it is an autonomous robot which detects the movement of the humans at the line of contact and it gives the exact location of the intruders and the live relay of the video will be sent to the server/base control station. This SD2 robot features both autonomous and manual control, it uses the PIR sensor for the identification of the human movement, pathfinding of the robot during autonomous control is provided using the 360° LIDAR, live relay of the video is provided using wide angle lens 5MP day and night camera, it has been provided with a GPS module for the live location of the surveillance area, and ubiquiti rocket and bullet technology is used for the communication purpose of the robot, this robot also features the self-powering technology using solar panel and self-destruction for the privacy and security purpose, all terrain design of the robot provides hassle-free surveillance. By using this featured SD2 robot, lives of the soldiers can be saved and the border surveillance is made easy and efficient.

Consumers nowadays have a higher expectation on cameras beyond the function of taking photos or videos. In the early years, the applications of cameras with lenses were, however, very restricted. A new generation of super wide-angle lenses capture the user’s surroundings in full 360°, allowing the user’s friends or families to step inside our universe and experience it, live. However, for such wide angle lens coverage, the relative illumination of the lens must take the vector nature of the light into account. Consequently, the polarization of light becomes a critical parameter. Custom coatings are commonly spread on the lens surface to avoid light transmission polarization falloff. But for consumer applications, complex coating is not a practical solution because it raises cost. In this paper, we present how to take care of the polarization during the design and analysis of the lens design. Additionally, design tips are proposed to suppress the polarization impact.

Advances in Optical Astronomical Instrumentation 2019; Melbourne; Australia; 9 December 2019 through 12 December 2019; Code 157181.--Proceedings of SPIE - The International Society for Optical Engineering Volume 11203, 2020, Article number 112031M, Entire Visible Sky (EnVisS) is one of the payload proposed for the ESA selected F-Class mission Comet Interceptor. The main aim of the mission is the study of a dynamically new comet, or an interstellar object, entering the inner solar system. The EnVisS camera is designed to capture the entire sky in some visible wavelength bands while the spacecraft passes through the comet's tail environment. EnVisS optical head is composed of a fisheye lens with a field of view of 180° x 40° coupled with an imaging detector equipped with both band-pass and polarimetric filters. Very wide angle lens, as a fisheye, must be necessarily anamorphic, i. e. its focal length must change along the field of view, in order to fit a finite-size imaging detector. This anamorphic distortion is introduced by the optical designer, depending on the desired applications. Each possible distortion bring along different field of view mapping and this must be taken into account by the scientific/metrological user, because the plate scale is variable along the focal plane. To obtain useful scientific information from fisheye images (astrometry, flux calibration and brightness measurements), a precise determination of the mapping function has to be accurately determined. In this paper we describe the expected distortion map of the EnVisS camera. © 2020 SPIE.

Visual perception is regularly used by humans and robots for navigation. By either implicitly or explicitly mapping the environment, ego-motion can be determined and a path of actions can be planned. The process of mapping and navigation are delicately intertwined; therefore, improving one can often lead to an improvement of the other. Both processes are sensitive to the interior orientation parameters of the camera system and mathematically modelling these systematic errors can often improve the precision and accuracy of the overall solution. This paper presents an automatic camera calibration method suitable for any lens, without having prior knowledge about the sensor. Statistical inference is performed to map the environment and localize the camera simultaneously. K-nearest neighbour regression is used to model the geometric distortions of the images. A normal-angle lens Nikon camera and wide-angle lens GoPro camera were calibrated using the proposed method, as well as the conventional bundle adjustment with self-calibration method (for comparison). Results showed that the mapping error was reduced from an average of 14.9 mm to 1.2 mm (i.e. a 92% improvement) and 66.6 mm to 1.5 mm (i.e. a 98% improvement) using the proposed method for the Nikon and GoPro cameras, respectively. In contrast, the conventional approach achieved an average 3D error of 0.9 mm (i.e. 94% improvement) and 3.3 mm (i.e. 95% improvement) for the Nikon and GoPro cameras, respectively. Thus, the proposed method performs well irrespective of the lens/sensor used: it yields results that are comparable to the conventional approach for normal-angle lens cameras, and it has the additional benefit of improving calibration results for wide-angle lens cameras., Comment: ISPRS TC I Mid-term Symposium "Innovative Sensing - From Sensors to Methods and Applications", 10-12 October 2018. Karlsruhe, Germany

In this research a human detection system is proposed in which people are viewed from an overhead camera with a wide angle lens. Due to perspective change a person can have different orientations and sizes at different positions in the scene relative to the optical centre. We exploit this property of the overhead camera and develop a novel algorithm which uses the variable size bounding boxes with different orientations, with respect to the radial distance of the center of the image. In these overhead view images we neither used any assumption about the pose or the visibility of a person nor imposed any restriction about the environment. When compare the results of proposed algorithm with a standard histogram of oriented gradient (HOG) algorithm, we achieve not only a huge gain in overall detection rate but also a significant improvement in reducing spurious detections per image. On average, 9 false detections occur per image. A new algorithm is proposed where transforming the image patch containing a person to remove positional dependency and then applying the HOG algorithm eliminates 98% of the spurious detections in noisy images from an industrial assembly line and detects people with a 95% efficiency.

We present a method for the automatic estimation of two-parameter radial distortion models, considering polynomial as well as division models. The method first detects the longest distorted lines within the image by applying the Hough transform enriched with a radial distortion parameter. From these lines, the first distortion parameter is estimated, then we initialize the second distortion parameter to zero and the two-parameter model is embedded into an iterative nonlinear optimization process to improve the estimation. This optimization aims at reducing the distance from the edge points to the lines, adjusting two distortion parameters as well as the coordinates of the center of distortion. Furthermore, this allows detecting more points belonging to the distorted lines, so that the Hough transform is iteratively repeated to extract a better set of lines until no improvement is achieved. We present some experiments on real images with significant distortion to show the ability of the proposed approach to automatically correct this type of distortion as well as a comparison between the polynomial and division models.

As computer based technologies are growing rapidly, new problems are arising, which need serious and urgent attentions. Person tracking is one of the typical computer vision problem which is the center of interest for researchers working on surveillance systems (industries, shopping malls, educational institutions and hospitals etc. In this research work, a top view camera has been installed with a wide angle lens able to cover a wide area and more information for surveillance and visual monitoring purposes. Some of the common issues that are addressed in this research work are occlusion, sudden change in movement, tracking standstill body, abrupt change in direction, varying lightening conditions and differentiating a person from other objects. In this paper, different tracking algorithms are compared on a newly developed dataset using OpenCV. These algorithms are pre-implemented in the popular OpenCV library. The results and efficiency of these algorithms on a new data set are also discussed. All these algorithms give different results on overhead and frontal view video sequences.

Through the Aspire high school internship program at the Johns Hopkins Applied Physics Laboratory, I have been able to learn hands on skills that have helped me determine the path for my future. Gaining exposure and the opportunity to collaborate in a technical professional environment has provided me with the valuable knowledge, engineering skills, and new learning approaches that augment my traditional academia. My assignment in the Geospace and Earth Science Group has been to repackage a camera lens for a mission proposal to study the Aurora Borealis from orbiting satellites. The mission goal supports space weather science in determining Earth’s interaction with surrounding space. To experiment with the engineering design process, I worked on the deconstruction of a practice lens. I determined the distances between the inner lenses to experiment designing and fabricating several lens holders. Through trial and error, I created the Version Three lens holder prototype that will be used in the design of the final wide-angle lens. After the final lens angle was calculated, two lenses with proper specifications were purchased. One of the lenses will be deconstructed and the other one will be used for a computerized tomography (CT) scan. This scan will provide the distances between the internal lenses that are necessary in the design of the final holders. This internship has allowed me to conduct research, consult experts, design and fabricate prototypes, learn time management skills, understanding the cost of off the shelf parts compared to the cost of labor, learning new technical skills, and networking.

Video surveillance is growing rapidly, new problems and issues are also coming into view which needs serious and urgent attention. Video surveillance system requires a beneficial energy efficient camera solution. In this paper, a single overhead camera solution is introduced which overcomes the problems ex-isting in various frontal and overhead based surveillance systems. This will increases the efficiency and accuracy of surveillance system i.e. frontal and overhead. In this paper, two energy efficient overhead camera models are presented. The first model consists of a single overhead camera with a wide angle lens which covers a wide field of view addresses problems present in the traditional surveillance system. The second model, presents a single smart centralized overhead camera which controls various frontal based cameras. Several factors associated with camera models such as field of view, focal length and distortion are also discussed. Impact of the surveillance cameras are finally discussed which shows that a single energy efficient overhead camera surveillance system can solves many problems present in traditional surveillance system like power consumption, storage, time, human resource and installation cost and small coverage area.

A very wide angle lens with a field of view of 360°x180° - a fisheye lens - has been designed to be used in a space environment. As a case study, the lens is assumed to be mounted on a spinning probe passing through a comet’s tail. The lens, rotating with the probe passing through the comet coma, may map the entire sky as viewed from the interior tail, providing unprecedented data on the spatial distribution of plasma and dust. Considering the foreseen space applications for the lens, radiation hardened glass has been taken into account for the design. A key feature of the lens is the “angular scale” uniformity (F-theta) of the sky distribution map projected on the focal plane allowing to obtain a reliable whole sky reconstruction. Care has also been taken to obtain an almost telecentric design, in order to permit filters placed on the focal plane to work properly. A telecentric fisheye operating with a pixel-limited resolution in the waveband from 500 nm up to 770 nm and with an F-theta distortion is presented in this paper.

The study proposes an outlier refinement methodology for automatic distortion rectification of wide-angle and fish-eye lens camera models in the context of streamlining vision-based tasks. The line-members sets are estimated in a scene through accumulation of line candidates emerging from the same edge source. An iterative optimization with an outlier refinement scheme was applied to the loss value, to simultaneously remove the extremely curved outliers from the line-members set and update the robust line members as well as estimating the best-fit distortion parameters with lowest possible loss. The proposed algorithm was able to rectify the distortions of wide-angle and fish-eye cameras even in extreme conditions such as heavy illumination changes and severe lens distortions. Experiments were conducted using various evaluation metrics both at the pixel-level (image quality, edge stretching effects, pixel-point error) as well as higher-level use-cases (object detection, height estimation) with respect to real and synthetic data from publicly available, privately acquired sources. The performance evaluations of the proposed algorithm have been investigated using an ablation study on various datasets in correspondence to the significance analysis of the refinement scheme and loss function. Several quantitative and qualitative comparisons were carried out on the proposed approach against various self-calibration approaches.

Here a refractometer with the liquid prism is proposed. It has a simple structure, low cost and a large measurable refractive index range. When the vertex angle of liquid prism is 30°, the measurable refractive index range is up to 1.00-3.86, while the common refractometers have only the measurable range of 1.30-1.70. The optical system of this refractometer is design, where the angle of field is 80° and the total length is 110mm. A linear array CCD is used as the image receiver. The design results show that MTF is 0.35 in the tangential surface and 0.55 in the sagittal surface at the Nyquist frequency of 70lp/mm and the full field. The MTF values of other fields are more than 0.6, and the maximum distortion is 0.37%.

This paper proposes Sneaking Detector, a system which recognizes sneaking on a laptop screen by other people and alerts the owner through several interventions. We utilize a pre-trained deep learning network to estimate eye gaze of sneakers captured by a front-facing camera. Since most of the cameras equipped on laptop computers cannot cover a wide enough range, a commercial wide-angle lens attachment and an image processing are applied in our system. On the dataset involving nine participants following four experiments, it has been realized that our system can estimate the horizontal eye gaze and recognizes whether a sneaker is looking at a screen or not with 78% accuracy.

International audience; This paper introduces an original video dataset containing dynamic index finger gestures in a mobile context. The dataset consists of 746 video sequences of 6 index finger gestures: left, right, up, down, tap, circle. The video sequences are obtained from a mobile phone's rear camera equipped with a wide angle lens, thus the dataset features particular challenges due to background motion and the distorted field of view. We present a baseline method that uses optical flow features in the form of a histogram to represent the motion information in the video sequences.

A novel method to evaluate lightning shield condition of catenaries system for electric railway using wideangle lens is proposed. This method realizes easy and simple evaluation of lightning risk of continual power transmission lines. Some results of fundamental trials in this paper as a first proposal of the method. According to these results, the possibility of lightning risk evaluation by image processing in the future is shown.