Hand-held Camera Research Articles

Playing the Flute in Shanghai: The Musical Life of Dai Shuhong

Playing the Flute in Shanghai: The Musical Life of Dai Shuhong

The Roses of Jay DeFeo and Bruce Conner: Belated Thoughts on Intermediality and Friendship

Bruce Conner’s 1967 short film, The White Rose, is an artifact of a particular place and time—Fillmore Street, San Francisco, one day in 1965—and a document of Conner’s friendship with the artist Jay DeFeo, painter of The Rose, the monumental painting at its center, which she labored over for eight years, and which had to be removed by a team of professionals when she was forced to give up her apartment and studio. Intermediality is a central property of Bruce Conner’s oeuvre, reflecting the background as visual artist he brought to filmmaking, beginning with A Movie in 1958. The poignant film he made years later of his friend and her work may be his most narrative but it is also profoundly intermedial, capturing with a handheld camera as gestural as a paint stroke, in a few minutes, the irrevocable spaces, textures, activities, and losses of the day The Rose left home.

Validating within-limb calibrated algorithm using a smartphone attached infrared thermal camera for detection of arthritis in children

ObjectivesTo determine the impact of physical activity on temperature after within-limb calibration (TAWiC) measures and their reproducibility. To determine if thermal imaging from a smartphone attached thermal camera is comparable to thermal imaging using a handheld thermal camera for detection of arthritis in children. MethodsChildren without symptoms were enrolled to the “asymptomatic exercise cohort”, and received infrared imaging, using a standard handheld camera, after initial resting period, after activity, and after second resting period. Children seen in the rheumatology clinic with knee pain were enrolled into the “symptomatic knee pain cohort” and received imaging with both the smartphone-attached and handheld cameras before a routine clinical exam. TAWiC was defined as the temperature differences between joint and ipsilateral mid-tibia as the main readout for arthritis detection. ResultsThe asymptomatic exercise cohort demonstrated notable changes in absolute and TAWiC temperatures collected by thermal imaging after physical activity, and temperatures did not consistently return to pre-activity levels after a second period of rest. The 95th TAWiC from anterior view were, resting one −0.1 C (0.5), activity −0.7 C (0.5), resting two −0.2 C (0.6) (resting 1 vs resting 2, p-value = 0.13). In the symptomatic knee pain cohort, the smartphone attached and handheld thermal cameras performed similarly in regards to detection of joint inflammation and evaluation of joint temperature using the TAWiC algorithm, with high sensitivity of 80% (55.2–100.0%) and specificity of 84.2% (76.0–92.4%) in the anterior knee view when compared with the gold standard joint exam performed by a pediatric rheumatologist. The mean 95th TAWiC temperature difference between the two cameras was −0.1 C (−0.1 to 0.0) (p = 0.0004). ConclusionsThis study showed continued validity of the TAWiC algorithm across two distinct thermal camera platforms and demonstrates promise for improved accessibility and utility of this technology for arthritis detection.

COMBINATION OF THERMAL INFRARED IMAGES AND LASERSCANNING DATA FOR 3D THERMAL POINT CLOUD GENERATION ON BUILDINGS AND TREES

Abstract. The thermal infrared study of urban environments is of growing interest. It allows to observe the variations of surface temperatures on objects over time and therefore the microclimate at the scale of a street. To facilitate the analysis of thermal interactions between urban elements, it is necessary to provide a 3D visualization of the thermography of a street. For this purpose, 3D thermal models combining geometric and thermal infrared (TIR) measurements are required. The chosen format for 3D thermal models is a point cloud with a temperature attribute. In our approach, two types of urban components are considered: buildings and trees. The geometric data of each component are acquired with a static laser scanner and the surface temperature is acquired with a thermal handheld camera. For the building, the approach consists in georeferencing TIR images and colorize the point cloud by projection. For trees, the approach consists of the colorization of each laser scan prior to the registration. The spherical panoramic images acquired with the Terrestrial Laser Scanning (TLS) are used as references to automatically georeferenced the TIR and thus to save time. The 3D thermal models obtained highlight the impact of sunlight on buildings and trees. At building scale, this thermal representation also helps to emphasize thermal bridges, as well as the shadow generated by surrounding trees. At tree scale, this representation is useful for monitoring the temporal and spatial variability of trunk’s and leave’s temperatures. Obviously, the thermal models underline the impact of trees on the urban environment.

Retinal vascular changes in response to hypoxia

AbstractPurpose: Increased tortuosity and engorgement of retinal vasculature are recognized physiological responses to hypoxia. This can lead to high‐altitude retinopathy (HAR), which is a pathology seen in unacclimatised individuals at altitude. It typically features asymptomatic flame haemorrhages, but incidence reports are highly variable and our understanding of the aetiological mechanisms remains incomplete. This study aimed to evaluate retinal vascular changes throughout an expedition to 4167 metres.Methods: 10 healthy participants summitted Mount Toubkal, Morocco. Fundus images were taken on a handheld camera pre‐departure, daily throughout the expedition, and one‐month post‐return. Diameter and tortuosity of four vessels was assessed, in addition to vessel density and the presence of HAR.Results: Significant (p ≤ 0.05) increases in tortuosity and diameter were observed in some vessels on high‐altitude exposure days. There was a significant increase in vessel density on summit day only.Conclusions: This is the first study to report increased vessel density and no incidence of HAR. These results are likely attributable to relatively low altitude exposure, a conservative ascent profile, and the young, healthy demographic profile of participants. However, the study is limited by its small sample size, environmental confounding factors and semi‐subjective diameter measurements. Physiological but not pathological changes were seen in this cohort, which gives insight to the state of the cerebral vasculature throughout this expedition and builds on current understanding of retinal vascular changes in hypoxia. Future work must include daily retinal images of larger sample sizes at higher altitude and take steps to mitigate against environmental confounders. This work is relevant to altitude tourists, patients with diabetic retinopathy or retinal vein occlusion, and critically ill patients at sea level.

Citrus green fruit detection via improved feature network extraction.

It is crucial to accurately determine the green fruit stage of citrus and formulate detailed fruit conservation and flower thinning plans to increase the yield of citrus. However, the color of citrus green fruits is similar to the background, which results in poor segmentation accuracy. At present, when deep learning and other technologies are applied in agriculture for crop yield estimation and picking tasks, the accuracy of recognition reaches 88%, and the area enclosed by the PR curve and the coordinate axis reaches 0.95, which basically meets the application requirements.To solve these problems, this study proposes a citrus green fruit detection method that is based on improved Mask-RCNN (Mask-Region Convolutional Neural Network) feature network extraction. First, the backbone networks are able to integrate low, medium and high level features and then perform end-to-end classification. They have excellent feature extraction capability for image classification tasks. Deep and shallow feature fusion is used to fuse the ResNet(Residual network) in the Mask-RCNN network. This strategy involves assembling multiple identical backbones using composite connections between adjacent backbones to form a more powerful backbone. This is helpful for increasing the amount of feature information that is extracted at each stage in the backbone network. Second, in neural networks, the feature map contains the feature information of the image, and the number of channels is positively related to the number of feature maps. The more channels, the more convolutional layers are needed, and the more computation is required, so a combined connection block is introduced to reduce the number of channels and improve the model accuracy. To test the method, a visual image dataset of citrus green fruits is collected and established through multisource channels such as handheld camera shooting and cloud platform acquisition. The performance of the improved citrus green fruit detection technology is compared with those of other detection methods on our dataset. The results show that compared with Mask-RCNN model, the average detection accuracy of the improved Mask-RCNN model is 95.36%, increased by 1.42%, and the area surrounded by precision-recall curve and coordinate axis is 0.9673, increased by 0.3%. This research is meaningful for reducing the effect of the image background on the detection accuracy and can provide a constructive reference for the intelligent production of citrus.

Diagnostic accuracy of handheld fundus photography: A comparative study of three commercially available cameras.

The objective of this study was to compare the sensitivity and specificity of handheld fundus cameras in detecting diabetic retinopathy (DR), diabetic macular edema (DME), and macular degeneration. Participants in the study, conducted at Maharaj Nakorn Hospital in Northern Thailand between September 2018 and May 2019, underwent an ophthalmologist examination as well as mydriatic fundus photography with three handheld fundus cameras (iNview, Peek Retina, Pictor Plus). Photographs were graded and adjudicated by masked ophthalmologists. Outcome measures included the sensitivity and specificity of each fundus camera for detecting DR, DME, and macular degeneration, relative to ophthalmologist examination. Fundus photographs of 355 eyes from 185 participants were captured with each of the three retinal cameras. Of the 355 eyes, 102 had DR, 71 had DME, and 89 had macular degeneration on ophthalmologist examination. The Pictor Plus was the most sensitive camera for each of the diseases (73-77%) and also achieved relatively high specificity (77-91%). The Peek Retina was the most specific (96-99%), although in part due to its low sensitivity (6-18%). The iNview had slightly lower estimates of sensitivity (55-72%) and specificity (86-90%) compared to the Pictor Plus. These findings demonstrated that the handheld cameras achieved high specificity but variable sensitivities in detecting DR, DME, and macular degeneration. The Pictor Plus, iNview, and Peek Retina would have distinct advantages and disadvantages when applied for utilization in tele-ophthalmology retinal screening programs.

Supporting workspace awareness in remote assistance through a flexible multi-camera system and Augmented Reality awareness cues

Workspace awareness is critical for remote assistance with physical tasks, yet it remains difficult to facilitate. For example, if the remote helper is limited to the single viewpoint provided by the worker’s hand-held or head-mounted camera, she lacks the ability to gain an overview of the workspace. This may be addressed by granting the helper view-independence, e.g., through a multi-camera system. However, it can be cumbersome to set up and calibrate multiple cameras, and it can be challenging for the local worker to identify the current viewpoint of the remote helper. We present CueCam, a multi-camera remote assistance system that supports mutual workspace awareness through a flexible ad-hoc camera calibration and various Augmented Reality cues that communicate the helper’s viewpoint and focus. In particular, we propose visual cues presented through a head-mounted Augmented Reality display (Virtual Hand, Color Cue), and sound cues emitted from the cameras’ physical locations (Spatial Sound). Findings from a lab study indicate that all proposed cues effectively support the worker’s awareness of helper’s location and focus, while the Color Cue demonstrated superiority in task performance and preference ratings during a search task.

Validity of Hand Held Fundus Camera by Optometrist using Slit lamp 90D bio microscopy as a reference standard for screening of Diabetes Retinopathy.

To find out the validity of hand-held fundus camera by optometrist using slit lamp 90D biomicroscopy as reference standard for screening of diabetes retinopathy. The observational cross-sectional study was conducted at the diabetic clinic of Al-Ibrahim Eye Hospital, Karachi, from August 2020 to May 2021, and comprised diabetics of either gender aged >16 years visiting the outpatient department. Un-dilated fundus photograph of both eyes were taken with non-mydriatic fundus camera. Pupils were then mid-dilated with one drop of tropicamide 1% before capturing retinal images by handheld fundus camera by another optometrist. Both the optometrists identified and recorded the presence and absence of diabetic retinopathy. Subsequently, a retinal specialist examined the fundus with slit lamp 90 D biomicroscopy. Data was analysed using SPSS 23. Of the 500 subjects, 291(58.2%) were males and 209(41.8%) were females. The overall mean age was 54.49±9.16 years (range: 16-83 years). Of the 1000 eyes, fundus was not readable in 130(13%) by hand-held fundus camera, 296(29.6%) eyes by non-mydriatic fundus camera and 76(7.6%) eyes by slit lamp. Sensitivity and specificity of hand-held fundus camera compared to non-mydriatic fundus camera was 89.86% and 80.36%, respectively. In comparison with slit lamp, the sensitivity was 91.71% and specificity was 71.10%. Kappa statistic for diabetic retinopathy detection by hand-held fundus camera versus non-mydriatic fundus camera was 0.705, indicating substantial agreement. Kappa statistic for diabetic retinopathy detection with hand-held fundus camera with semi-dilated pupil was found to be a valid screening tool in the hand of optometrist for preliminary screening of diabetic retinopathy. Handheld fundus camera with semi-dilated pupil was found to be a valid screening tool in the hand of optometrist for preliminary screening of diabetic retinopathy.

Victorian Facebooks: Privacy Concerns at William Notman’s Studio

The history of issues of privacy in relation to photography is generally dated to the introduction of handheld cameras and halftone printing at the end of the nineteenth century. Citing a remarkable series of notations in the records of famed Montreal photographer William Notman, this article argues that privacy concerns were at issue in the everyday operations of the photography studios that preceded amateur photography. These notations represent the rudimentary attempts that some of Notman’s clients made to restrict the circulation, sale and availability of their studio portraits starting in the 1870s. This article speculates that the emergent anxieties regarding privacy exemplified by such notations were informed by the setup and marketing practices of early photography studios. Ultimately, this analysis seeks to show that current debates about privacy, photography and visibility can be traced back to the commercial systems that developed in tandem with early photography technologies.

Lightweight and Parameter-Optimized Real-Time Food Calorie Estimation from Images Using CNN-Based Approach

Automated object identification has seen significant progress during the last decade with close to human-level accuracy, aided by deep learning methods. With the rapid rise of obesity and other lifestyle-related diseases worldwide, the availability of fast, automated, and reliable image-based food calorie estimation is becoming a necessity. With the help of a deep learning-based automated object identification system, it is possible to introduce accurate and intelligent solutions in the form of a mobile app. However, for these kind of applications, processing speed is an important concern as the images should be processed in real time. Although plenty of studies have been conducted that focus on food image detection-based calorie estimation, there is still a lack of an image-driven, lightweight, fast, and reliable food calorie estimation system. In this paper, we propose a method based on the parameter-optimized Convolution Neural Networks (CNN) for detecting food images of regular meals using a handheld camera. Once identification process of the food items are complete, the corresponding calories and nutritional facts can be calculated using prior knowledge about the food class. Through our findings, we demonstrate that our proposed approach ensures high accuracy and can significantly simplify the existing manual calorie estimation procedures by converting them into a real-time automated process.

Land Use Land Cover Labeling of GLOBE Images Using a Deep Learning Fusion Model.

Most of the land use land cover classification methods presented in the literature have been conducted using satellite remote sensing images. High-resolution aerial imagery is now being used for land cover classification. The Global Learning and Observations to Benefit, the Environment land cover image database, is created by citizen scientists worldwide who use their handheld cameras to take a set of six images per land cover site. These images have clutter due to man-made objects, and the pixel uncertainties result in incorrect labels. The problem of accurate labeling of these land cover images is addressed. An integrated architecture that combines Unet and DeepLabV3 for initial segmentation, followed by a weighted fusion model that combines the segmentation labels, is presented. The land cover images with labels are used for training the deep learning models. The fusion model combines the labels of five images taken from the north, south, east, west, and down directions to assign a unique label to the image sets. 2916 GLOBE images have been labeled with land cover classes using the integrated model with minimal human-in-the-loop annotation. The validation step shows that our architecture of labeling the images results in 90.97% label accuracy. Our fusion model can be used for labeling large databases of land cover classes from RGB images.

Hand-held and UAV camera comparison in building thermal inspection process

Thermal building inspections indicate the growing interest among researchers due to the global climate changes resulting in legislations and policy contribution to low-energy consumption. The technologies applied during the experiments involve infrared cameras: both hand-held and on unmanned aerial vehicles. The research conducted in the public building in Kielce included the analysis of glass partition from the inside of the building by means of hand-held and UAV camera. The temperature distributions presented in the graphs proved that both methods are accurate in case of such investigations. However, during the experiments, it was concluded that hand-held camera is far more convenient.

A data driven approach to generate realistic 3D tree barks

3D models of trees are ubiquitous in video games, movies, and simulators. It is of paramount importance to generate high quality 3D models to enhance the visual content, and increase the diversity of the available models. In this work, we propose a methodology to create realistic 3D models of tree barks from a consumer-grade hand-held camera. Additionally, we present a pipeline that makes use of multi-view 3D Reconstruction and Generative Adversarial Networks (GANs) to generate the 3D models of the barks. We introduce a GAN referred to as the Depth-Reinforced-SPADE to generate the surfaces of the tree barks and the bark color concurrently. This GAN gives extensive control on what is being generated on the bark: moss, lichen, scars, etc. Finally, by testing our pipeline on different Northern-European trees whose barks exhibit radically different color patterns and surfaces, we show that our pipeline can be used to generate a broad panel of tree species’ bark.

Speed, Direction, Color and Type Identification of NHAI Expansions Using Deep Learning

Abstract: Vehicle numeration is associate interaction to appraise the road traffic thickness to judge the traffic conditions for shrewd transportation frameworks. With the broad use of cameras in metropolitan vehicle frameworks, the reconnaissance mission video has become a focal info supply to boot, constant traffic the board framework has become illustrious as lately owing to the accessibility of handheld/versatile cameras and machine learning investigation. In this work, propose video-based vehicle as well as technique in associate superhighway traffic video caught utilizing hand-held cameras. The primary and therefore the necessary step to estimate the vehicle flow; this later helps North American nation to count the vehicles victimization the virtual line Generally, we have a tendency to begin with the background subtraction to isolate moving objects. To facilitate crossing of vehicles with the road, we have a tendency to apply the detection of objects. Our system uses the LBPH (Local Binary Pattern Histogram) algorithmic program as a way to deduct the background, so as to use our numeration algorithmic program. Traffic observance is one space that utilizes Deep Learning for many functions. By exploitation cameras put in in some spots on the roads, several tasks like vehicle investigating, vehicle identification, traffic violation observance, vehicle speed observance, etc. will be completed. Deep Learning may be a common Machine Learning formula that's wide employed in several areas in current way of life. Its strong performance and ready-to-use frameworks and architectures allows many of us to develop varied Deep Learning-based code or systems to support human tasks and activities. During this paper, we tend to discuss a Deep Learning implementation to form a vehicle investigating system while not having to trace the vehicles movements. to reinforce the system performance and to cut back time in deploying Deep Learning design, therefore pre-trained model of YOLOv3 is employed during this analysis because of its sensible performance and moderate process time in object detection. This analysis aims to form a straightforward vehicle investigating system to assist human in classify and investigating the vehicles that cross the road. The investigating relies on four varieties of vehicle, i.e. car, motorcycle, bus, and truck, whereas previous analysis counts the automobile solely. because the result, our planned system capable to count the vehicles crossing the road supported video captured by camera with the very best accuracy of ninety seven.97%.

An ISM-BN-GA based methodology for product emotional design

Most current studies of product emotional design (PED) focus on the association of emotion image with a single design feature. However, these studies have ignored the fact that all product features have a combined effect on emotion image, and have also ignored the relationship between the whole and the part. In view of this, starting from the systematicness of design, PED is translated to the solution of the PED system. Moreover, an ISM-BN-GA-based methodology for PED is proposed. First, the sense perception–perceptual cognition–perceptual synthesis–emotional decision (PCSD) cognitive model is developed based on the process of user emotion transmission, based on which a pedigree of the elements in the PED system is constructed. Second, a multilevel hierarchical directed network of the elements in the PED system is established using the interpretation structure model (ISM) system engineering tool, and the network topology is extracted as the network structure of a Bayesian network (BN). Then, the BN is trained via parameter learning, and the evaluation model of the PED system based on the ISM-BN is constructed. Third, with the ISM-BN evaluation model as the objective function, a genetic algorithm (GA) is used to search for the optimal design scheme. Taking the design of a handheld camera as an example, the validity of the proposed method is verified, which proves its feasibility and practicability in the field of PED.

Volumetric assessment of hepatic grafts using a light detection and ranging system for 3D scanning: Preliminary data.

BACKGROUNDLiver transplantation has evolved into a safe life-saving operation and remains the golden standard in the treatment of end stage liver disease. The main limiting factor in the application of liver transplantation is graft shortage. Many strategies have been developed in order to alleviate graft shortage, such as living donor partial liver transplantation and split liver transplantation for adult and pediatric patients. In these strategies, liver volume assessment is of paramount importance, as size mismatch can have severe consequences in the success of liver transplantation.AIMTo evaluate the safety, feasibility, and accuracy of light detection and ranging (LIDAR) 3D photography in the prediction of whole liver graft volume and mass.METHODSSeven liver grafts procured for orthotopic liver transplantation from brain deceased donors were prospectively measured with an LIDAR handheld camera and their mass was calculated and compared to their actual weight.RESULTSThe mean error of all measurements was 17.03 g (range 3.56-59.33 g). Statistical analysis of the data yielded a Pearson correlation coefficient index of 0.9968, indicating a strong correlation between the values and a Student’s t-test P value of 0.26. Mean accuracy of the measurements was calculated at 97.88%.CONCLUSIONOur preliminary data indicate that LIDAR scanning of liver grafts is a safe, cost-effective, and feasible method of ex vivo determination of whole liver volume and mass. More data are needed to determine the precision and accuracy of this method.

A decision support tool for shelf-life determination of strawberries using hyperspectral imaging technology

Due to high perishability and poor distribution management, strawberry is one of the most frequently discarded fruits. The aim of this study was to develop a non-destructive system for accurate estimation of shelf-life of strawberries using hyperspectral imaging technology. Harvested strawberries were stored for 9 day at five different temperatures. Shelf-life was calculated based on subjective visual evaluation of appearance attributes (colour, shrivelling, and decay) using a rating scale. Hyperspectral images of strawberries were obtained during cold storage, using a novel handheld push broom line-scanning hyperspectral camera. The model developed by partial least square regression (PLSR) with selected spectra was used to predict appearance scores of strawberries with a coefficient of determination of prediction (R 2 p) of 0.97 and root mean square error of prediction (RMSEp) of 0.17. The appearance scores from the PLSR model were used to develop a model based on first-order kinetics and Arrhenius equations to predict the remaining shelf-life of the strawberries. The models predicted remaining shelf-life with R 2 p of 0.86 and RMSEp of 1.4 days. Prediction models were also developed for other quality attributes and biochemical properties (weight loss, ascorbic acid, and soluble solids). The results from this study support the development a non-destructive system for the accurate estimation of shelf-life of strawberries. A system that could potentially offer objective quality assessment as product moves through the supply chain, aiding decision making and facilitating proactive actions to be taken with the aim of minimising loss/waste; benefiting food supply chain stakeholders across the globe and supporting the drive for zero waste. • Establishing a non-destructive system for shelf-life estimation of strawberries. • Remaining shelf-life of strawberries determined along the supply chain. • Novel system supports for reducing waste along strawberry supply chain.

High dynamic range and super-resolution from raw image bursts

Photographs captured by smartphones and mid-range cameras have limited spatial resolution and dynamic range, with noisy response in underexposed regions and color artefacts in saturated areas. This paper introduces the first approach (to the best of our knowledge) to the reconstruction of highresolution, high-dynamic range color images from raw photographic bursts captured by a handheld camera with exposure bracketing. This method uses a physically-accurate model of image formation to combine an iterative optimization algorithm for solving the corresponding inverse problem with a learned image representation for robust alignment and a learned natural image prior. The proposed algorithm is fast, with low memory requirements compared to state-of-the-art learning-based approaches to image restoration, and features that are learned end to end from synthetic yet realistic data. Extensive experiments demonstrate its excellent performance with super-resolution factors of up to ×4 on real photographs taken in the wild with hand-held cameras, and high robustness to low-light conditions, noise, camera shake, and moderate object motion.

Singing a Great Dream: The Revolutionary Songs and Life of Khusiram Pakhrin

Singing a Great Dream: The Revolutionary Songs and Life of Khusiram Pakhrin