Faster R-CNN for Object Location in a Virtual Environment for Sorting Task

Pinzón-Arenas, Javier Orlando; Jimenez, Robinson; Murillo, Paula C. Useche

doi:10.3991/ijoe.v14i07.8465

Cited by 6 publications

(3 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…is user group can accommodate thousands of users at the same time [15]. (3) Real-time can also be regarded as dynamic [16]:…”

Section: Basketball Equipmentmentioning

confidence: 99%

Distributed Virtual Environment Basketball Equipment Embedded Systems’ Research and Development

Zhou

2021

Mathematical Problems in Engineering

View full text Add to dashboard Cite

With the development of social economy and the improvement of material level, people are paying more and more attention to their own health, and they are also constantly improving their fitness awareness. As far as the current sports equipment is concerned, the boring exercise methods can no longer satisfy people’s pursuit of higher quality fitness exercises. The embedding of the distributed virtual environment can incorporate its perceptual, immersive, and interactive characteristics, which increases the realization of multiuser interactive behavior in the field of sports equipment in the virtual scene. Therefore, this paper proposes the research and development of basketball sports equipment embedded system based on distributed virtual environment. First of all, this article adopts the literature method to learn the application of distributed virtual environment and embedded system in depth and, secondly, designs the embedded system architecture of basketball sports equipment suitable for distributed virtual environment and the software and hardware framework of the embedded system. Finally, the performance, delay comparison, system optimization test, and other aspects of the embedded system are analyzed. The average delay of interactive functions of embedded systems based on distributed virtual environment is 1.4 min, which saves 4 min compared with traditional interactive methods. And, the throughput rate of the system has also been greatly improved. The research on the embedded system of sports equipment based on the distributed virtual environment proposed in this paper will be beneficial to the development of society and economy, and it can also help more people to enrich the way of sports and enhance the fun of sports.

show abstract

“…is user group can accommodate thousands of users at the same time [15]. (3) Real-time can also be regarded as dynamic [16]:…”

Section: Basketball Equipmentmentioning

confidence: 99%

Distributed Virtual Environment Basketball Equipment Embedded Systems’ Research and Development

Zhou

2021

Mathematical Problems in Engineering

View full text Add to dashboard Cite

show abstract

“…This strategy can be implemented on an exoskeleton to restore stable walking in individuals with paralysis caused by SCI [11]. Moreover, a convolutional neural network was used to aid a mobile robotic arm in the process of object detection and classification [12]. Other researchers in the field designed an adaptive fuzzy controller to control a robotic arm based on oscillator and differentiator.…”

Section: Introductionmentioning

confidence: 99%

Control of an Exoskeleton for Lower Limb Rehabilitation Using ANFIS

Arabiat¹,

Matahen²,

Zaid³

et al. 2022

Int. J. Onl. Eng.

View full text Add to dashboard Cite

Exoskeletons are powered robotic devices designed to be worn by humans to provide physical assistance or power augmentation. In this work, a control system for a powered exoskeleton is designed. This exoskeleton is aimed at aiding in the rehabilitation of Spinal Bifidas. Spinal Bifida is the most common disability in childhood after Cerebral Palsy, it is a defective development of the spinal cord during conception. Two phases for this work are presented: system identification and control using ANFIS. While it is difficult to attain an accurate dynamical model of complex system, this work employed ANFIS to help control and stabilize the system. Gait trajectories were obtained by modeling the system as a linear inverted pendulum, a simulation was performed with a traditional controller. Afterwards, trajectory data was obtained and used to train and test ANFIS to create the model and controller. One, two and three inputs were investigated to train the ANFIS. Results showed that the one-input model visibly failed to follow the trajectory. The average RMSE for the two-input model was 0.096, and for the three-inputs, the RMSE on average was higher; 0.19, making it worse, however the knee model contrastingly showed improvement, as the RMSE was lower by 2% for the knee specifically.

show abstract

“…The second part is Fast R-CNN which is used to sort proposals. Faster R-CNN has 9 anchors consisting of 3 scales and 3 ratios that make this method can detect objects more accurately [11]- [13]. When we use R-CNN, the bounding boxes (BBs) are generated [14].…”

mentioning

confidence: 99%

Comparison of two deep learning methods for detecting fire hotspots

Lestari

Kosasih

2022

IJECE

View full text Add to dashboard Cite

Every high-rise building must meet construction requirements, i.e. it must have good safety to prevent unexpected events such as fire incident. To avoid the occurrence of a bigger fire, surveillance using closed circuit television (CCTV) videos is necessary. However, it is impossible for security forces to monitor for a full day. One of the methods that can be used to help security forces is deep learning method. In this study, we use two deep learning methods to detect fire hotspots, i.e. you only look once (YOLO) method and faster region-based convolutional neural network (faster R-CNN) method. The first stage, we collected 100 image data (70 training data and 30 test data). The next stage is model training which aims to make the model can recognize fire. Later, we calculate precision, recall, accuracy, and F1 score to measure performance of model. If the F1 score is close to 1, then the balance is optimal. In our experiment results, we found that YOLO has a precision is 100%, recall is 54.54%, accuracy is 66.67%, and F1 score is 0.70583667. While faster R-CNN has a precision is 87.5%, recall is 95.45%, accuracy is 86.67%, and F1 score is 0.913022.

show abstract

Faster R-CNN for Object Location in a Virtual Environment for Sorting Task

Cited by 6 publications

References 14 publications

Distributed Virtual Environment Basketball Equipment Embedded Systems’ Research and Development

Distributed Virtual Environment Basketball Equipment Embedded Systems’ Research and Development

Control of an Exoskeleton for Lower Limb Rehabilitation Using ANFIS

Comparison of two deep learning methods for detecting fire hotspots

Contact Info

Product

Resources

About