Rear Camera Research Articles

Mobile Phone-Based Use of the Photoplethysmography Technique to Detect Atrial Fibrillation in Primary Care: Diagnostic Accuracy Study of the FibriCheck App.

BackgroundMobile phone apps using photoplethysmography (PPG) technology through their built-in camera are becoming an attractive alternative for atrial fibrillation (AF) screening because of their low cost, convenience, and broad accessibility. However, some important questions concerning their diagnostic accuracy remain to be answered.ObjectiveThis study tested the diagnostic accuracy of the FibriCheck AF algorithm for the detection of AF on the basis of mobile phone PPG and single-lead electrocardiography (ECG) signals.MethodsA convenience sample of patients aged 65 years and above, with or without a known history of AF, was recruited from 17 primary care facilities. Patients with an active pacemaker rhythm were excluded. A PPG signal was obtained with the rear camera of an iPhone 5S. Simultaneously, a single‑lead ECG was registered using a dermal patch with a wireless connection to the same mobile phone. PPG and single-lead ECG signals were analyzed using the FibriCheck AF algorithm. At the same time, a 12‑lead ECG was obtained and interpreted offline by independent cardiologists to determine the presence of AF.ResultsA total of 45.7% (102/223) subjects were having AF. PPG signal quality was sufficient for analysis in 93% and single‑lead ECG quality was sufficient in 94% of the participants. After removing insufficient quality measurements, the sensitivity and specificity were 96% (95% CI 89%-99%) and 97% (95% CI 91%-99%) for the PPG signal versus 95% (95% CI 88%-98%) and 97% (95% CI 91%-99%) for the single‑lead ECG, respectively. False-positive results were mainly because of premature ectopic beats. PPG and single‑lead ECG techniques yielded adequate signal quality in 196 subjects and a similar diagnosis in 98.0% (192/196) subjects.ConclusionsThe FibriCheck AF algorithm can accurately detect AF on the basis of mobile phone PPG and single-lead ECG signals in a primary care convenience sample.

On the identities of Rhinoleucophenga pallida Hendel and Rhinoleucophenga obesa (Loew) (Diptera, Drosophilidae), with description of a new sibling species from Brazil

Rhinoleucophenga pallidaHendel, 1917 (type species of the genus) is redescribed based on its female holotype and a male from a nearby locality, and Rhinoleucophenga obesa (Loew, 1872) on its two syntypes, which are designated as the male lectotype and a female paralectotype. Both are valid species. A proposal is made to establish the genus Pseudophortica Sturtevant, 1918 (type species R. obesa), a junior synonym of Rhinoleucophenga, to subgenus rank and include all species of Rhinoleucophenga described or redescribed from males except R. pallida, which is unique in having a remarkable pedunculate surstylus, among other differences. The North American R. obesa is compared to its closest sibling, the South American species Rhinoleucophenga gigantea (Thomson, 1869). The occurrence of R. obesa in Brazil is also questioned, as suggested long ago by Marshall R. Wheeler. The specimens from Brazil previously identified as such most probably belong to the new species described in the present paper as Rhinoleucophenga (Pseudophortica) cantareira sp. nov. (type locality: Parque Estadual da Cantareira, City of São Paulo, State of São Paulo, Brazil). Numerous photomicrographs of their habitus and male terminalia taken with a Smartphone's rear camera and digitally stacked to create images with greater depth of focus are provided.

I-VALS: Visual Attention Localization for Mobile Service Computing

Identifying and localizing the user's visual attention can enable various intelligent service computing paradigms in a mobile environment. However, existing solutions can only compute the gaze direction, but without the distance to the intended target. In addition, most of them rely on eye tracker or similar infrastructure support. This paper explores the possibility of using portal mobile devices, e.g., smartphone, to detect the visual attention of a user. i-VALS only requires the user to do one simple action to localize the intended object: gazing at the intended object and holding up the smartphone so that the object and the user's face can be simultaneously captured by the front and rear cameras. We develop efficient algorithms to obtain both the distance between the camera and user, the user's gaze direction and the object's direction from the camera. The object's location can then be computed by solving a trigonometric problem. i-VALS has been prototyped on commercial off-the-shelf (COTS) devices. The extensive experiment results show that i-VALS achieves high accuracy and small latency, effectively supporting a large variety of applications in smart environments.

Smartphone-Based Diagnosis of Parasitic Infections With Colorimetric Assays in Centrifuge Tubes

A smartphone-based platform for the diagnosis of parasitic infections has been developed, tested and validated. The system is capable of making automatic and accurate analysis of millimetric colorimetric arrays in centrifuge collection tubes, which are well established tools used in clinical analysis. To that end, an Android-based software application has been developed, making use of the smartphone rear camera, enabling precise image processing of the colorimetric spot arrays. A low-cost plastic accessory has been developed using 3D-printing to provide controlled illumination, fixed sample positioning and cell phone attachment. The platform was then tested repeatedly for its size detection, edge blurriness and colour detection capabilities. A minimum spot radius of $175~\mu \text{m}$ is detectable when using the developed app, with a tolerance of 15%, corresponding to 0.25% of the area where the spot array is printed. Spot edge definition has been studied up to 40% of blurriness, resulting in a low average percentage error of 1.24%. Colour detection follows the well-known Gamma correction function. Finally, the whole platform was tested and validated using real DNA to analyse for accurate discrimination of Trypanosomatid species, which are responsible for devastating diseases in humans and livestock. The smartphone-based platform can be further extended to other clinical analysis. Its simplicity and reliable performance mean it can be used in remote, limited-resource settings by relatively unskilled technicians/nurses, where diagnostic laboratories are sparsely distributed. The results can however be sent easily via the smartphone to medical experts as well as government health agencies.

Real-world effects of rear automatic braking and other backing assistance systems

Objective: To evaluate the effects of rearview cameras, rear parking sensors, and rear automatic braking systems on backing crashes. Method: Negative binomial regression was used to compare police-reported backing crash involvements per insured vehicle year in 23 US states during 2012–2015 among General Motors vehicles with Rear Vision Camera alone; Rear Parking Assist alone (rear parking sensors); Rear Vision Camera and Rear Parking Assist; or the Rear Vision Camera, Rear Parking Assist, and Rear Automatic Braking to vehicles with none of these systems. Modeling controlled for other backing assistance systems on vehicles and factors that may affect crash risk. Results: The combination of Rear Vision Camera and Rear Parking Assist reduced backing crash involvement rates by 42%. When Rear Automatic Braking was added to the Rear Vision Camera and Rear Parking Assist, vehicles with all three systems had backing crash involvement rates that were 78% lower than vehicles with none of the systems. On vehicles with Rear Parking Assist alone or Rear Vision Camera alone, backing crash involvement rates were reduced 28% and 5%, respectively, but these reductions were not statistically significant. Conclusions: Rearview cameras and rear parking sensors are preventing some backing crashes, but their effectiveness may be constrained in part by drivers not using or responding to the systems appropriately. Rear automatic braking adds to the effectiveness of these systems because it does not rely entirely on appropriate driver response. Practical applications: Rear parking sensors and rearview cameras are available on most new vehicles, but availability of rear automatic braking is limited. If more vehicles were equipped with rear automatic braking that performed like the system evaluated in the current study, many backing crashes that still occur among vehicles with rearview cameras and rear parking sensors could be prevented.

Learning Scene Illumination by Pairwise Photos from Rear and Front Mobile Cameras

AbstractIllumination estimation is an essential problem in computer vision, graphics and augmented reality. In this paper, we propose a learning based method to recover low‐frequency scene illumination represented as spherical harmonic (SH) functions by pairwise photos from rear and front cameras on mobile devices. An end‐to‐end deep convolutional neural network (CNN) structure is designed to process images on symmetric views and predict SH coefficients. We introduce a novel Render Loss to improve the rendering quality of the predicted illumination. A high quality high dynamic range (HDR) panoramic image dataset was developed for training and evaluation. Experiments show that our model produces visually and quantitatively superior results compared to the state‐of‐the‐arts. Moreover, our method is practical for mobile‐based applications.

Penerapan Augmented Reality Berbasis Android Sebagai Bahan Ajar Untuk Pemodelan Mesin Injection

The value obtained by Fadilah Vocational School students about motorcycle injection systems is still below the value of mastery learning. Many factors cause students to get low scores in the competence to understand the gasoline fuel system, especially the Honda motorcycle injection system. One of the factors that makes it difficult for students to understand injection material is the unavailability of Honda PGM-FI system props. The limited number of teaching aids is a shortage of private schools in general due to the limited costs of procuring teaching aids for the PGM-FI system. Less interesting learning model is also a factor in the low value of students in certain subjects. Augmented Reality is a developing technology in the world, combining two-dimensional or three-dimensional virtual objects into a real three-dimensional environment and then projecting these virtual objects in real time. Three dimensions commonly abbreviated as 3D, or space, are forms of objects that have length, width, and height. This term is usually used in the fields of art, animation, computers and mathematics. The application of Augmented Reality is not like virtual reality that completely replaces reality, but only adds or completes reality by using libraryVuforia, the image in the form of three dimensions will appear on the screen of the Android smartphone assisted by the rear camera. The application of Augmented Reality in learning is more interesting because this technology provides real visualization. So that learning in the field of education is more interesting and uses the latest technology so that students are more interested in learning.

LOCALIZATION OF A CAR BASED ON MULTI-SENSOR FUSION

Abstract. The vehicle localization is an essential component for stable autonomous car operation. There are many algorithms for the vehicle localization. However, it still needs much improvement in terms of its accuracy and cost. In this paper, sensor fusion based localization algorithm is used for solving this problem. Our sensor system is composed of in-vehicle sensors, GPS and vision sensors. The localization algorithm is based on extended Kalman filter and it has time update step and measurement update step. In the time update step, in-vehicle sensors are used such as yaw-rate and speed sensor. And GPS and vision sensor information are used to update the vehicle position in the measurement update step.We use visual odometry library to process vision sensor data and generate the moving distance and direction of the car. Especially, when performing visual odometry we use georeferenced image database to reduce the error accumulation. Through the experiments, the proposed localization algorithm is verified and evaluated. The RMS errors of the estimated result from the proposed algorithm are about 4.3 m. This result shows about 40 % improvement in accuracy even in comparison with the result from the GPS only method. It shows the possibility to use proposed localization algorithm. However, it is still necessary to improve the accuracy for applying this algorithm to the autonomous car. Therefore, we plan to use multiple cameras (rear cameras or AVM cameras) and more information such as high-definition map or V2X communication. And the filter and error modelling also need to be changed for the better results.

INITIAL EVALUATION OF 3D RECONSTRUCTION OF CLOSE OBJECTS WITH SMARTPHONE STEREO VISION

Abstract. The Worldwide spread of relatively low cost mobile devices embedded with dual rear cameras enables the possibility of exploiting smartphone stereo vision for producing 3D models. Despite such idea is quite attractive, the small baseline between the two cameras restricts the depth discrimination ability of this kind of stereo vision systems. This paper presents the results obtained with a smartphone stereo vision system by using two rear cameras with different focal length: this operating condition clearly reduces the matchable area. Nevertheless, 3D reconstruction is still possible and the obtained results are evaluated for several camera-object distances.

Automatic Calibration of an Around View Monitor System Exploiting Lane Markings.

This paper proposes a method that automatically calibrates four cameras of an around view monitor (AVM) system in a natural driving situation. The proposed method estimates orientation angles of four cameras composing the AVM system, and assumes that their locations and intrinsic parameters are known in advance. This method utilizes lane markings because they exist in almost all on-road situations and appear across images of adjacent cameras. It starts by detecting lane markings from images captured by four cameras of the AVM system in a cost-effective manner. False lane markings are rejected by analyzing the statistical properties of the detected lane markings. Once the correct lane markings are sufficiently gathered, this method first calibrates the front and rear cameras, and then calibrates the left and right cameras with the help of the calibration results of the front and rear cameras. This two-step approach is essential because side cameras cannot be fully calibrated by themselves, due to insufficient lane marking information. After this initial calibration, this method collects corresponding lane markings appearing across images of adjacent cameras and simultaneously refines the initial calibration results of four cameras to obtain seamless AVM images. In the case of a long image sequence, this method conducts the camera calibration multiple times, and then selects the medoid as the final result to reduce computational resources and dependency on a specific place. In the experiment, the proposed method was quantitatively and qualitatively evaluated in various real driving situations and showed promising results.

Path-Following Control Based on Ground-Watching Navigation

This paper discusses the path-following performance of actively-steered articulated vehicles based on measurement signals from two “ground-watching” navigation systems (GWNSs). The GWNSs are described. These uses high-speed USB-3 cameras and an image processing computer to measure and calculate off-tracking distance for a path-following steering controller. Full-scale field tests are performed using an experimental articulated vehicle, tested under both open-loop and closed-loop conditions. Off-tracking distance at the rear camera with respect to the front camera is controlled to be less than 0.1 m for tests with both GWNSs.

Path-Following Control Based on Ground-Watching Navigation

This paper discusses the path-following performance of actively-steered articulated vehicles based on measurement signals from two “ground-watching” navigation systems (GWNSs). The GWNSs are described. These uses high-speed USB-3 cameras and an image processing computer to measure and calculate off-tracking distance for a path-following steering controller. Full-scale field tests are performed using an experimental articulated vehicle, tested under both open-loop and closed-loop conditions. Off-tracking distance at the rear camera with respect to the front camera is controlled to be less than 0.1 m for tests with both GWNSs.

Use of Phone Sensors to Enhance Distracted Pedestrians’ Safety

Studies have shown that using smartphones while walking—called distracted walking —significantly increases the risk of pedestrians colliding with dangerous objects. In this paper, we explore how to mitigate this problem by exploiting the phone's built-in sensors only and developing an application called BumpAlert . BumpAlert provides a generic solution without requiring any prior knowledge of the user's surroundings by estimating distances to nearby objects using the phone's speakers and microphones. This process is enhanced further by using the images acquired from the phone's rear camera, when necessary. We have evaluated BumpAlert under a variety of settings ranging from aisle to outdoor environments with walls, pillars, signboards, dustbins, and people, etc., that are common in our daily surroundings. Our evaluation has shown an average accuracy of BumpAlert to be higher than 95 percent with a less than 2 percent false-positive rate to detect frontal objects 2–4m away, which suffices for the user to react and avoid collision. Even though BumpAlert is unable to detect all dangerous situations, most participants of our user study feel safer when they walk with BumpAlert enabled. Integrating our current design of BumpAlert with other safety systems can provide a practical solution for protecting distracted pedestrians.

Non-Invasive Oxygen Determination in Intelligent Packaging Using a Smartphone

Here, we present a technique for the determination of the gaseous oxygen concentration inside packed food. It is based on the use of a luminescent membrane sensitive to O2 that is optically excited and read by a smartphone. The flash of the smartphone along with an optical filter is used as the light source for the optical stimulation of the membrane. The luminescence generated, which is quenched by the surrounding gaseous oxygen, is registered by the rear camera of the same device. The response parameter is defined by combining the registered intensities at two different wavelength ranges corresponding to the emission and the absorption peaks of the sensitive membrane. Thanks to this novel response parameter, the sensitivity is increased and, more importantly, the thermal dependence of the membrane is significantly reduced. This approach allows the use of a luminescent O2-sensitive membrane for intelligent packaging with no need of any associated electronics for its excitation and reading. This means that an oxygen sensor is developed, where a luminescent compound acts as an indicator, therefore combining the advantages of both schemes, that is, the simplicity and reduced cost of indicators with the high sensitivity and accuracy of selective sensors.

Collaborative approach for a safe driving distance using stereoscopic image processing

Disregard for the rules regarding the minimum safety distance can make the avoidance of a rear-end collision nearly impossible. In a joint effort to enhance safety and improve the decision making processes on an individual level, we contribute to the state of the art with an innovative and affordable system that identifies vehicles and provides a rear-end distance warning system capable of recognizing dangerous situations, and which can also inform other vehicles of the danger, independent of their communication capabilities or equipment. Vision sensors garner information through the stereoscopic capturing and processing of images by rear cameras to calculate the distance between the leading and following vehicles. Visual data related to the safety distance is provided to the following vehicle in real-time, relying on an asynchronous collaborative process. A detailed error analysis of the distance calculation is provided based on the measurement procedure and roadway geometry. Relying on the communication between the two vehicles, an in-vehicle system was compared to the rear-mounted distance warning system under lab-controlled conditions. Both human–machine interaction paradigms were evaluated in terms of their impact on driver response. Results showed that both systems influenced the driver in keeping a time gap of two seconds.

Advanced Safety Vehicle (ASV) Technology Driver Support System Monitor using Three Onboard Camera

This paper presents the development of safe driving support system using three onboard cameras. One camera monitors the driver to determine the current attention location, and the other is a front camera that detects pedestrians, running lane and vehicles in front. The rear camera detects pedestrians and approaching vehicles. The pedestrians and vehicles are detected using a specially trained HOG/Adaboost system. Lane detections use edge detection and RANSAC. Information from the three cameras is then used to determine if a situation is dangerous enough to warrant warning the driver. We have conducted experiments, and the results confirm that this system has the potential to support automatic driving.

Finger gesture input utilizing the rear camera of a mobile phone: A perspective of mobile CAD

AbstractVarious interfaces have been suggested for mobile devices, including touch gesture and embedded sensor‐based interfaces. However, if a user's task requires a thorough look, these interfaces hinder sight of view in display and thus can be inappropriate for the tasks. This problem is more important especially in the mobile computer aided design (CAD) context, which performs visually demanding tasks on the limited screen. To address this point, this study suggests a mobile interface utilizing the rear camera of the device and describes the function mapping for CAD. The suggested interface detects finger attachment and movement direction. The number of colors in camera vision and local binary pattern are used as features, and a support vector machine (SVM) is used for feature classification. A prototype application is designed for validation, and appropriate SVM models are selected through benchmarking tests. The validation results show that the suggested interface can perform with high accuracy and low computational resources.

Augmented reality in laser laboratories

Laser safety glasses block visibility of the laser light. This is a big nuisance when a clear view of the beam path is required. A headset made up of a smartphone and a viewer can overcome this problem. The user looks at the image of the real world on the cellphone display, captured by its rear camera. An unimpeded and safe sight of the laser beam is then achieved.If the infrared blocking filter of the smartphone camera is removed, the spectral sensitivity of the CMOS image sensor extends in the near infrared region up to 1100 nm. This substantial improvement widens the usability of the device to many laser systems for industrial and medical applications, which are located in this spectral region.The paper describes this modification of a phone camera to extend its sensitivity beyond the visible and make a true augmented reality laser viewer.

A system for classifying vegetative structures on coffee branches based on videos recorded in the field by a mobile device

An expert system development to classify six coffee vegetative structures.A non-destructive system based on a smartphone camera and computer vision techniques.A system robust to illumination and occlusion changes using 2D and 3D features. As a drink, coffee is one of the most in demand products worldwide; as an agricultural product, it requires non-destructive tools for its monitoring and control. In order to create a non-destructive method which can be used in the field, a system was developed to find and classify six types of vegetative structures on coffee branches: leaves, stems, flowers, unripe fruits, semi-ripe fruits, and ripe fruits. Videos were obtained from 12 coffee branches in field conditions, using the rear camera of a mobile device. Approximately 90 frames, those which had the most information from the scene, were selected from each video. Next, a three-dimensional (3D) reconstruction was generated using the Structure from Motion (SfM) and Patch-based Multi-view Stereo (PMVS) techniques for each branch. All acquired images were manually recorded, and a Ground Truth point cloud was generated for each branch. The generated point clouds were filtered using a statistical outliers filter, in order to eliminate noise generated in the 3D reconstruction process. The points that were located in the deepest part were considered to be the scene background, and were removed using a band-pass filter. Point clouds were sub-sampled using a VoxelGrid filter, to reduce the number of points to 50% and therefore reduce computation time of the processes that followed. Various two-dimensional (2D) and 3D features were taken from the point clouds: 11 based on RGB, Lab, Luv, YCbCr, and HSV color space, four based on curvatures, and the remaining two based on shape and curvedness indexes. A Support Vector Machine (SVM) was trained with the previously mentioned features by using eight branches for the training stage, and four branches for the validation stage. Experimental results showed a precision of 0.82 and a recall of 0.79, when classifying said vegetative structures. The proposed system is economical, as only a mobile device is needed to obtain information. Remaining system processes were performed offline. Additionally, the system developed was not affected by changes in lighting conditions, when recording videos on a coffee plantation.

Enhancing GPS With Lane-Level Navigation to Facilitate Highway Driving

Lane-level navigation has received a lot of attention in recent years. It has played a great role in assisting route planning, as well as navigating automated vehicles. Aside from sticking to the planned route, abnormal traffic situations which result in blocking lanes could impact lane switching decisions. Unfortunately, there is currently no navigation system that can sense and track a vehicle's lane position and to advise the driver of lane switching decisions. Google Maps stores a priori the number of lanes and their directions at each highway exit and provides this information to drivers when navigating. However, even with this information, some drivers may not be able to make an informed decision regarding when and where to make a correct lane switch. This motivated us to develop a mechanism for the detection and tracking of real-time lane changes. In this paper, we propose a GPS-aiding system that can sense and track a vehicle's lane position. The system leverages smart phones’ computing capability, rear cameras, and inertial motion sensors. With little extra computational overhead, the system applies computer vision techniques to achieve lane-level positioning. We also design a machine learning-based algorithm to detect and track lane switching. We conduct a series of experiments, analyze our system in real-world environments, and achieve very promising results. We believe our system can be a great asset to current smart phone navigation systems.