scholarly journals Automated Quantification of Macronutrients using Computer Vision on a Depth-Sensing Smartphone (Preprint)

2019 ◽  
Author(s):  
David Herzig ◽  
Christos T Nakas ◽  
Janine Stalder ◽  
Christophe Kosinski ◽  
Céline Laesser ◽  
...  
Keyword(s):  
Computer Vision ◽  
Processing Time ◽  
Energy Content ◽  
Absolute Error ◽  
Smartphone Application ◽  
Estimation Accuracy ◽  
Depth Sensor ◽  
Depth Sensing ◽  
Food Items ◽  
Macronutrient Content

BACKGROUND Quantification of dietary intake is key to the prevention and management of numerous metabolic disorders. Conventional approaches are challenging, laborious, and, suffer from lack of accuracy. The recent advent of depth-sensing smartphones in conjunction with computer vision has the potential to facilitate reliable quantification of food intake. OBJECTIVE To evaluate the accuracy of a novel smartphone application combining depth-sensing hardware with computer vision to quantify meal macronutrient content. METHODS The application ran on a smartphone with built-in depth sensor applying structured light (iPhone X) and estimated weight, macronutrient (carbohydrate, protein, fat) and energy content of 48 randomly chosen meals (type of meals: breakfast, cooked meals, snacks) encompassing 128 food items. Reference weight was generated by weighing individual food items using a precision scale. The study endpoints were fourfold: i) error of estimated meal weight; ii) error of estimated meal macronutrient content and energy content; iii) segmentation performance; and iv) processing time. RESULTS Mean±SD absolute error of the application’s estimate was 35.1±42.8g (14.0±12.2%) for weight, 5.5±5.1g (14.8±10.9%) for carbohydrate content, 2.4±5.6g (13.0±13.8%), 1.3±1.7g (12.3±12.8%) for fat content and 41.2±42.5kcal (12.7±10.8%) for energy content. While estimation accuracy was not affected by the viewing angle, the type of meal mattered with slightly worse performance for cooked meals compared to breakfast and snack. Segmentation required adjustment for 7 out of 128 items. Mean±SD processing time across all meals was 22.9±8.6s. CONCLUSIONS The present study evaluated the accuracy of a novel smartphone application with integrated depth-sensing camera and found a high accuracy in food estimation across all macronutrients. This was paralleled by a high segmentation performance and low processing time corroborating the high usability of this system.

scholarly journals BLAINDER—A Blender AI Add-On for Generation of Semantically Labeled Depth-Sensing Data

Sensors ◽  
2021 ◽  
Vol 21 (6) ◽  
pp. 2144
Author(s):  
Stefan Reitmann ◽  
Lorenzo Neumann ◽  
Bernhard Jung
Keyword(s):  
Point Clouds ◽  
Training Data ◽  
Sensor Data ◽  
Generation Process ◽  
Data Generation ◽  
Depth Sensor ◽  
Depth Sensing ◽  
Depth Sensors ◽  
3D Point Clouds ◽  
Wide Range

Common Machine-Learning (ML) approaches for scene classification require a large amount of training data. However, for classification of depth sensor data, in contrast to image data, relatively few databases are publicly available and manual generation of semantically labeled 3D point clouds is an even more time-consuming task. To simplify the training data generation process for a wide range of domains, we have developed the BLAINDER add-on package for the open-source 3D modeling software Blender, which enables a largely automated generation of semantically annotated point-cloud data in virtual 3D environments. In this paper, we focus on classical depth-sensing techniques Light Detection and Ranging (LiDAR) and Sound Navigation and Ranging (Sonar). Within the BLAINDER add-on, different depth sensors can be loaded from presets, customized sensors can be implemented and different environmental conditions (e.g., influence of rain, dust) can be simulated. The semantically labeled data can be exported to various 2D and 3D formats and are thus optimized for different ML applications and visualizations. In addition, semantically labeled images can be exported using the rendering functionalities of Blender.

scholarly journals Development and Usability Assessment of a Food Record App for Dietary Assessments among Older Adults with Type 2 Diabetes (Preprint)

2019 ◽  
Author(s):  
Hyunggu Jung ◽  
George Demiris ◽  
Peter Tarczy-Hornoch ◽  
Mark Zachry
Keyword(s):  
Older Adults ◽  
Qualitative Data ◽  
Dietary Assessment ◽  
Smartphone Application ◽  
Design Principles ◽  
Assessment Methods ◽  
Food Record ◽  
Food Items ◽  
Usability Assessment ◽  
Dietary Assessment Methods

BACKGROUND More than one in four people in the United States aged 65 years and older have diabetes. For diabetes care, medical nutrition therapy (MNT) is recommended as a clinically effective intervention. Prior researchers have developed and validated dietary assessment methods using images of food items for improving the accuracy of self-reporting over traditional methods. Nevertheless, little is known about the usability of image-assisted dietary assessment methods for older adults with diabetes. OBJECTIVE The aims of this study were: a) to create a food record app for dietary assessments (FRADA) that would support image-assisted dietary assessments, and b) to evaluate the usability of FRADA for older adults with diabetes. METHODS For the development of FRADA, we identified design principles that address the needs of older adults and implemented three fundamental tasks required for image-assisted dietary assessments: capturing, viewing, and transmitting images of food based on the design principles. For the usability assessment of FRADA, older adults aged 65 to 80 (11 females and 3 males) were assigned to interact with FRADA in a lab-based setting. Participants’ opinions of FRADA and its usability were determined by a follow-up survey and interview. As an evaluation indicator of usability, the responses to the survey including an After-Scenario Questionnaire were analyzed. Qualitative data from the interviews confirmed the responses to the survey. RESULTS We developed a smartphone application that enables older adults with diabetes to capture, view, and transmit images of food items they consumed. The findings of this study showed that FRADA and its instructions for capturing, viewing, and transmitting images of food items were usable for older adults with diabetes. The survey showed that FRADA was easy to use, and study participants would consider using FRADA daily. The analysis of the qualitative data from interviews revealed multiple themes, such as the usability of FRADA, potential benefits and features of FRADA, and concerns of older adults with diabetes when interacting with FRADA. CONCLUSIONS This study demonstrates in a lab-based setting, not only the usability of FRADA with older adults who have diabetes, but it also demonstrates potential opportunities using FRADA in real-life settings. The findings suggest implications for creating a smartphone application for an image-assisted dietary assessment. Future work still remains to evaluate the feasibility and validity of FRADA with multiple stakeholders involving older adults with diabetes and dietitians.

scholarly journals Real-Time Monitoring of Jet Trajectory during Jetting Based on Near-Field Computer Vision

Sensors ◽  
2019 ◽  
Vol 19 (3) ◽  
pp. 690 ◽  
Author(s):  
Jinsong Zhu ◽  
Wei Li ◽  
Da Lin ◽  
Ge Zhao
Keyword(s):  
Computer Vision ◽  
Prediction Model ◽  
Real Time ◽  
Processing Time ◽  
Near Field ◽  
Image Correction ◽  
Noise Elimination ◽  
Sensor Device ◽  
Jet Trajectory ◽  
Novel Method

A novel method of near-field computer vision (NFCV) was developed to monitor the jet trajectory during the jetting process, which was used to precisely predict the falling point position of the jet trajectory. By means of a high-resolution webcam, the NFCV sensor device collected near-field images of the jet trajectory. Preprocessing of collected images was carried out, which included squint image correction, noise elimination, and jet trajectory extraction. The features of the jet trajectory in the processed image were extracted, including: start-point slope (SPS), end-point slope (EPS), and overall trajectory slope (OTS) based on the proposed mean position method. A multiple regression jet trajectory range prediction model was established based on these trajectory characteristics and the reliability of the model was verified. The results show that the accuracy of the prediction model is not less than 94% and the processing time is less than 0.88s, which satisfy the requirements of real-time online jet trajectory monitoring.

Attitude and Position Estimation in UAVs using Artificial Landmarks and MEMS Sensors in a Virtual Environment

2020 ◽  
Author(s):  
Rafael Costa Fernandes ◽  
Paulo Sergio Silva ◽  
Felipe Ieda Fazanaro ◽  
Diego Paolo Ferruzzo Correa
Keyword(s):  
Computer Vision ◽  
Processing Time ◽  
Estimation Algorithm ◽  
Position Estimation ◽  
Simulation Software ◽  
Estimation Technique ◽  
Mems Sensors ◽  
Microelectromechanical System ◽  
Hybrid Estimation ◽  
The Cost

This work discusses the development of a hybrid estimation algorithm based on computer vision and microelectromechanical system sensors. A mathematical enviroment was developed to simulate the dynamics of the quadrotor and its sensors, a 3D simulation software was also developed, simulating a on-board camera. The results obtained were compared to a TRIAD/MEMS attitude and position estimation technique. A fourty times increase in precision was shown, at the cost of five times additional computational processing time.

Estimating Putting Outcomes in Golf: Experts Have a Better Sense of Distance

2018 ◽  
Vol 125 (2) ◽  
pp. 313-328 ◽  
Author(s):  
Hideyuki Tanaka ◽  
Masato Iwami
Keyword(s):  
Motion Capture ◽  
Partial Correlation ◽  
Control Variable ◽  
Absolute Error ◽  
Estimation Accuracy ◽  
Motion Capture System ◽  
Mixed Design ◽  
Specific Training ◽  
Lower Variability ◽  
The Relationship

In putting, golfers require an internal forward sense of the causal relationship between putting actions and outcomes—a sense of distance—to decide appropriate impact intensity. As no previous work has shown such a cognitive ability in skilled golfers, we sought to quantify sense-of-distance skill differences between experts and novice golfers in both putting-swing consistency and accuracy of outcome estimation. We compared nine expert and nine novice golfers on putting-outcome estimation by having them putt a golf ball to a target located at three distances (1.2, 2.4, and 3.6 m), and then, after automatic closure of their electric-shutter spectacles immediately following putter impact with the ball, they gave their best estimate of where the ball stopped. We assessed outcome-estimation accuracy by calculating the absolute error between the stopped ball’s actual and estimated positions. We also measured and analyzed putter head-swing movements during the task using a motion-capture system. Two-way, mixed-design analysis of variance tests revealed that expert golfers achieved both significantly lower variability in putter-head kinematics and higher accuracy at outcome estimation than the novices. Linear partial correlation analyses with target distance as the control variable tested the relationship between outcome-estimation performance and putter-head variability kinematic measurements. There were no significant correlations between them for experts and novices in separate databases, while medium correlations were found in a collective database. Thus, swing consistency and a sense of distance are independent skills that both account for putting expertise, and specific training is required for each to improve putting skills.

scholarly journals Computer Vision Based Virtual Sketch Using Detection

2022 ◽  
Vol 10 (1) ◽  
pp. 264-268
Author(s):  
Santosh Dhaigude
Keyword(s):  
Computer Vision ◽  
Online Learning ◽  
Processing Time ◽  
Recognition Accuracy ◽  
New Techniques ◽  
Research Areas ◽  
Controlling System ◽  
Learning Path ◽  
Keyword Detection ◽  
Challenging Research

Abstract: In todays world during this pandemic situation Online Learning is the only source where one could learn. Online learning makes students more curious about the knowledge and so they decide their learning path . But considering the academics as they have to pass the course or exam given, they need to take time to study, and have to be disciplined about their dedication. And there are many barriers for Online learning as well. Students are lowering their grasping power the reason for this is that each and every student was used to rely on their teacher and offline classes. Virtual writing and controlling system is challenging research areas in field of image processing and pattern recognition in the recent years. It contributes extremely to the advancement of an automation process and can improve the interface between man and machine in numerous applications. Several research works have been focusing on new techniques and methods that would reduce the processing time while providing higher recognition accuracy. Given the real time webcam data, this jambord like python application uses OpenCV library to track an object-of-interest (a human palm/finger in this case) and allows the user to draw bymoving the finger, which makes it both awesome and interesting to draw simple thing. Keyword: Detection, Handlandmark , Keypoints, Computer vision, OpenCV

scholarly journals Development of Hand Gesture Based Electronic Key Using Microsoft Kinect

2018 ◽  
Vol 218 ◽  
pp. 02014
Author(s):  
Arief Ramadhani ◽  
Achmad Rizal ◽  
Erwin Susanto
Keyword(s):  
Computer Vision ◽  
Gesture Recognition ◽  
Recognition System ◽  
Hand Gesture Recognition ◽  
Microsoft Kinect ◽  
Hand Gesture ◽  
Depth Sensor ◽  
Average Accuracy ◽  
Recognition Result ◽  
Hand Image

Computer vision is one of the fields of research that can be applied in a various subject. One application of computer vision is the hand gesture recognition system. The hand gesture is one of the ways to interact with computers or machines. In this study, hand gesture recognition was used as a password for electronic key systems. The hand gesture recognition in this study utilized the depth sensor in Microsoft Kinect Xbox 360. Depth sensor captured the hand image and segmented using a threshold. By scanning each pixel, we detected the thumb and the number of other fingers that open. The hand gesture recognition result was used as a password to unlock the electronic key. This system could recognize nine types of hand gesture represent number 1, 2, 3, 4, 5, 6, 7, 8, and 9. The average accuracy of the hand gesture recognition system was 97.78% for one single hand sign and 86.5% as password of three hand signs.

scholarly journals A multicentre study of street foods purchased in urban areas of Central Asia: the FEEDCities Project

2019 ◽  
Vol 29 (Supplement_4) ◽  
Author(s):  
S Sousa ◽  
N Lunet ◽  
M Gelormini ◽  
J Jewell ◽  
I Morais ◽  
...  
Keyword(s):  
Public Health ◽  
Central Asia ◽  
Urban Areas ◽  
Energy Content ◽  
Nutritional Composition ◽  
Systematic Sampling ◽  
Trans Fat ◽  
Street Food ◽  
Food Items ◽  
Food Composition Tables

Abstract Background Street food (SF) is a strong tradition in Central Asia, where urbanization and westernization of food habits is occurring. Research on SF consumption is scarce, and crucial to understand its implications for public health. This study aims to describe the SF purchases in urban areas of Tajikistan, Kyrgyzstan, Turkmenistan and Kazakhstan. Methods A cross-sectional study was conducted in Dushanbe, Bishkek, Ashgabat and Almaty in 2016/2017. SF markets (n = 34) and vending sites (n = 270) were selected by random and systematic sampling. Data on customers’ characteristics and food items purchased was collected by direct observation. Nutritional composition of the food items (n = 852) was obtained by laboratorial analysis (n = 582) or food composition tables and labels (n = 270). Results A total of 714 customers were identified. The most commonly purchased foods and beverages were savoury pastries/snacks (23.2%), main dishes (19.0%), sweet pastries/confectionery (17.9%), tea/coffee (11.3%) and soft drinks/juices (9.8%). Fruit was the least frequently purchased food (1.1%). Nearly one-third of customers purchased industrial food items; this proportion was significantly higher in Kazakhstan (43.2%) and Turkmenistan (32.3%). The median energy content of a SF purchase ranged between 352kcal (Tajikistan) and 568kcal (Turkmenistan). The median saturated (SFA) and trans fat contents were 4.74g and 0.36g, respectively; the highest values were 9.01g for SFA (Turkmenistan) and 0.60g for trans fat (Kazakhstan), accounting for 40.6% and 27.3% of the maximum daily recommendations, respectively. Sodium-potassium ratio was far above recommended, reaching the highest values of 6.57 and 5.17 in Tajikistan and Kyrgyzstan. Conclusions Frequent purchase of industrial food reflects a shift to a westernized dietary pattern. Public health policies in these settings should aim to increase fruit availability and to improve SF nutritional composition, namely its lipid profile and sodium content. Key messages A relevant proportion of customers bought industrial foods, while fruit was rarely purchased, reflecting the nutrition transition process that is occurring in developing countries. Street food meals showed concerning levels of saturated fat, trans-fat and sodium, which must be considered when designing strategies targeted to improve the urban food environment in these settings.

scholarly journals Perceptions of portion size and energy content: implications for strategies to affect behaviour change

2011 ◽  
Vol 15 (2) ◽  
pp. 246-253 ◽  
Author(s):  
Emily Brindal ◽  
Carlene Wilson ◽  
Philip Mohr ◽  
Gary Wittert
Keyword(s):  
Fast Food ◽  
Energy Content ◽  
Portion Size ◽  
Average Error ◽  
Cross Sectional ◽  
Energy Estimation ◽  
Food Items ◽  
Computer Based ◽  
Males And Females ◽  
Actual Size

AbstractObjectiveTo assess Australian consumers’ perception of portion size of fast-food items and their ability to estimate energy content.DesignCross-sectional computer-based survey.SettingAustralia.SubjectsFast-food consumers (168 male, 324 female) were asked to recall the items eaten at the most recent visit to a fast-food restaurant, rate the prospective satiety and estimate the energy content of seven fast-food or ‘standard’ meals relative to a 9000 kJ Guideline Daily Amount. Nine dietitians also completed the energy estimation task.ResultsRatings of prospective satiety generally aligned with the actual size of the meals and indicated that consumers perceived all meals to provide an adequate amount of food, although this differed by gender. The magnitude of the error in energy estimation by consumers was three to ten times that of the dietitians. In both males and females, the average error in energy estimation for the fast-food meals (females: mean 3911 (sd 1998) kJ; males: mean 3382 (sd 1957) kJ) was significantly (P < 0·001) larger than for the standard meals (females: mean 2607 (sd 1623) kJ; males: mean 2754 (sd 1652) kJ). In women, error in energy estimation for fast-food items predicted actual energy intake from fast-food items (β = 0·16, P < 0·01).ConclusionsKnowledge of the energy content of standard and fast-food meals in fast-food consumers in Australia is poor. Awareness of dietary energy should be a focus of health promotion if nutrition information, in its current format, is going to alter behaviour.

Sign in / Sign up

Export Citation Format

Share Document