Investigations of Object Detection in Images/Videos Using Various Deep Learning Techniques and Embedded Platforms—A Comprehensive Review

Murthy, Chinthakindi Balaram; Hashmi, Mohammad Farukh; Bokde, Neeraj Dhanraj; Geem, Zong Woo

doi:10.3390/app10093280

Cited by 100 publications

(31 citation statements)

References 195 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Labeled datasets such as PASCAL VOC dataset [ 19 ] and Common Objects in Context (COCO) [ 20 ] have played important roles in the continuous improvement of 2D detection systems. Nice reviews of 2D detection systems can be found in [ 21 , 22 ].…”

Section: Related Workmentioning

confidence: 99%

3D Object Detection and Instance Segmentation from 3D Range and 2D Color Images

Shen

Stamos

2021

Sensors

View full text Add to dashboard Cite

Instance segmentation and object detection are significant problems in the fields of computer vision and robotics. We address those problems by proposing a novel object segmentation and detection system. First, we detect 2D objects based on RGB, depth only, or RGB-D images. A 3D convolutional-based system, named Frustum VoxNet, is proposed. This system generates frustums from 2D detection results, proposes 3D candidate voxelized images for each frustum, and uses a 3D convolutional neural network (CNN) based on these candidates voxelized images to perform the 3D instance segmentation and object detection. Results on the SUN RGB-D dataset show that our RGB-D-based system’s 3D inference is much faster than state-of-the-art methods, without a significant loss of accuracy. At the same time, we can provide segmentation and detection results using depth only images, with accuracy comparable to RGB-D-based systems. This is important since our methods can also work well in low lighting conditions, or with sensors that do not acquire RGB images. Finally, the use of segmentation as part of our pipeline increases detection accuracy, while providing at the same time 3D instance segmentation.

show abstract

Section: Related Workmentioning

confidence: 99%

3D Object Detection and Instance Segmentation from 3D Range and 2D Color Images

Shen

Stamos

2021

Sensors

View full text Add to dashboard Cite

show abstract

“…At the same time, it presents better performance when combining with other techniques. A comprehensive study for all the presentation attack detection methods is discussed for the face recognition systems in [28,29]. One of these solutions use a light field camera to reveal the presentation attack [30].…”

Section: Inputmentioning

confidence: 99%

Detection and Recognition of Moving Objects for Physical Systems Protection in Nuclear Facilities

Gaber

Khalaf

Mahmoud

et al. 2021

Journal of Advanced Engineering Trends

View full text Add to dashboard Cite

Recently, the nuclear applications and services are entering in several domains of our life. It is a mandatory direction to protect the physical systems against any malicious sabotage or threat. Human activity detection and recognition techniques have a significant role in the physical protection systems. It enhances the protection regime through anomalous activity detection and authorized human-computer interaction applications. This paper is aimed to demonstrate a real-time system for moving object\human detection and recognition in order to protect the physical systems. It integrates computer vision with Internet of Things (IoT) technologies for detecting and recognizing the moving humans with their occupation times at the camera field. The proposed system can introduce more reliable and active intrusion detection scenarios with simple and low cost implementation techniques. It provides high flexibility for securing access to the monitoring information with storing for the historian purposes. Through an internet environment, the system can monitor the sensitive areas through detecting the unknown faces and their occupation times via ThingSpeak channels. The monitoring information is tracked through web based dashboards and ThingView mobile application. The system can launch automatically the alert notifications through SMS, and e-mail messages. These notifications are received whenever detecting an intruder face and exceeding the occupation times of human motion. The results provide effective features for implementing real-time and low cost applications even with low resolution cameras.

show abstract

“…It is good at detecting tiny and meaningful spatial features and has the characteristics of sparse weights [55]. In general, CNN is applied to computer vision areas, such as image classification [56] and object detection [57]. Therefore, it is also suitable for the vision-based slab arching detection problem in this paper.…”

Section: Cnn For Slab Arching Detectionmentioning

confidence: 99%

Arching Detection Method of Slab Track in High-Speed Railway Based on Track Geometry Data

Gao

Zhong

et al. 2020

Applied Sciences

View full text Add to dashboard Cite

During the long-term service of slab track, various external factors (such as complicated temperature) can result in a series of slab damages. Among them, slab arching changes the structural mechanical properties, deteriorates the track geometry conditions, and even threatens the operation of trains. Therefore, it is necessary to detect slab arching accurately to achieve effective maintenance. However, the current damage detection methods cannot satisfy high accuracy and low cost simultaneously, making it difficult to achieve large-scale and efficient arching detection. To this end, this paper proposed a vision-based arching detection method using track geometry data. The main works include: (1) data nonlinear deviation correction and arching characteristics analysis; (2) data conversion and augmentation; (3) design and experiments of convolutional neural network- based detection model. The results show that the proposed method can detect arching damages effectively, and the F1-score reaches 98.4%. By balancing the sample size of each pattern, the performance can be further improved. Moreover, the method outperforms the plain deep learning network. In practice, the proposed method can be employed to detect slab arching and help to make maintenance plans. The method can also be applied to the data-based detection of other structural damages and has broad prospects.

show abstract

Investigations of Object Detection in Images/Videos Using Various Deep Learning Techniques and Embedded Platforms—A Comprehensive Review

Cited by 100 publications

References 195 publications

3D Object Detection and Instance Segmentation from 3D Range and 2D Color Images

3D Object Detection and Instance Segmentation from 3D Range and 2D Color Images

Detection and Recognition of Moving Objects for Physical Systems Protection in Nuclear Facilities

Arching Detection Method of Slab Track in High-Speed Railway Based on Track Geometry Data

Contact Info

Product

Resources

About