Comparison of the YOLOv3 and Mask R-CNN architectures’ efficiency in the smart refrigerator’s computer vision

Доррер, М. Г.; Tolmacheva, A. E.

doi:10.1088/1742-6596/1679/4/042022

Cited by 15 publications

(7 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In [28], Mask R-CNN architecture is compared with YOLOv3. The dataset contains 800 training and 70 test images It was found that the accuracy of Mask R-CNN is significantly higher compared to YOLOv3, but in terms of detection speed, YOLOv3 outperformed Mask R-CNN.…”

Section: Introduction and Related Workmentioning

confidence: 99%

Comparing YOLOv3, YOLOv4 and YOLOv5 for Autonomous Landing Spot Detection in Faulty UAVs

Nepal

Eslamiat

2022

Sensors

270

View full text Add to dashboard Cite

In-flight system failure is one of the major safety concerns in the operation of unmanned aerial vehicles (UAVs) in urban environments. To address this concern, a safety framework consisting of following three main tasks can be utilized: (1) Monitoring health of the UAV and detecting failures, (2) Finding potential safe landing spots in case a critical failure is detected in step 1, and (3) Steering the UAV to a safe landing spot found in step 2. In this paper, we specifically look at the second task, where we investigate the feasibility of utilizing object detection methods to spot safe landing spots in case the UAV suffers an in-flight failure. Particularly, we investigate different versions of the YOLO objection detection method and compare their performances for the specific application of detecting a safe landing location for a UAV that has suffered an in-flight failure. We compare the performance of YOLOv3, YOLOv4, and YOLOv5l while training them by a large aerial image dataset called DOTA in a Personal Computer (PC) and also a Companion Computer (CC). We plan to use the chosen algorithm on a CC that can be attached to a UAV, and the PC is used to verify the trends that we see between the algorithms on the CC. We confirm the feasibility of utilizing these algorithms for effective emergency landing spot detection and report their accuracy and speed for that specific application. Our investigation also shows that the YOLOv5l algorithm outperforms YOLOv4 and YOLOv3 in terms of accuracy of detection while maintaining a slightly slower inference speed.

show abstract

Section: Introduction and Related Workmentioning

confidence: 99%

Comparing YOLOv3, YOLOv4 and YOLOv5 for Autonomous Landing Spot Detection in Faulty UAVs

Nepal

Eslamiat

2022

Sensors

270

View full text Add to dashboard Cite

show abstract

“…In contrast, the increase in multiplicity reaches three percent, indicating that the category imbalance loss function can more effectively distinguish hard-to-discriminate targets. 7 that it has a higher detection accuracy compared with the common two-stage detection algorithms Faster R-CNN [12] and Mask R-CNN [13]. Compared with the one-stage detection algorithms YOLOv2 [14], YOLOv3 [15], and SSD [16] series, it highlights its own detection accuracy more.…”

Section: Experiments and Analysis Of Resultsmentioning

confidence: 94%

Research on Mini-EfficientDet Identification Algorithm Based on Transfer Learning

Liu

Shao

2022

J. Phys.: Conf. Ser.

View full text Add to dashboard Cite

In this research paper, the Chinese beehive culture in Fujian Province, China is used as the carrier, and the Mini-EfficientDet deep neural network after migration is used to identify common species at the hive door, that is, the identification of Chinese bees, wasps and cockroaches in the form of nymphs. In this paper, we define the modified model as Mini-EfficientDet by compressing the initial EfficientDet model and adding the category imbalance function, which makes it focus more on the recognition and classification of small targets while ensuring the recognition accuracy. Through the test on the MSCOCO2017 data set, it is concluded that when the backbone network adopts EfficientNet B7, it shows strong detection accuracy in detecting targets of various scales, which confirms the role of the category imbalance function proposed in this paper and the efficiency of the improved EfficientDet model. Detection accuracy. The pre-trained model is transferred to the field of beehive species detection through migration learning, that is, after the post-training of the self-collected data set, the detection accuracy of Chinese bee, cockroach, and wasp is 98.66%, 83.71%, and 82.06%. It has made sufficient algorithmic preparations for the later detection and early warning system of beehive species invasion.

show abstract

“…Our study in "III. PROBLEMS ENCOUNTERED AND POSSIBLE SOLUTIONS" shows that for the existing state-of-the-art Tracking-by-Detection [15,[26][27][28] and Target tracking [44][45][46][47] methods available are either too low in precision and recall rate or too slow to be directly applied to the autonomous landing of fixed-wing UAVs. Therefore, we propose a deep learning based Vision Transformer Particle Region-based Convolutional Neural Network (VitP-RCNN).…”

Section: Discussionmentioning

confidence: 99%

“…This study concluded that YOLOv3 has higher mAP and FPS than Faster RCNN and SSD. In Dorrer [28], Mask RCNN architecture was compared with YOLOv3. It was found that the accuracy of Mask RCNN is significantly higher compared to YOLOv3, but in terms of detection speed, YOLOv3 outperformed Mask RCNN.…”

Section: Trackingbydetection Methodsmentioning

confidence: 99%

High Speed Safe Autonomous Landing Marker Tracking of Fixed Wing Drone Based on Deep Learning

Yuan

Wang

2022

IEEE Access

View full text Add to dashboard Cite

Most of the current research on autonomous landing of Unmanned Aerial Vehicles (UAVs) focus on rotorcrafts, which can fly horizontally over the landing site and then land vertically, with less landing risk. However, some of the unmanned fixed-wing aircrafts take off and land in the sliding mode of wheel landing gears, and their landing stage is difficult to control and easier to failure. In this paper, the autonomous landing of the fixed wing UAVs based on the visual navigation is studied. We propose a deep learning based Vision Transformer Particle Region-based Convolutional Neural Network (VitP-RCNN). Our VitP-RCNN uses Mobile Vision Transformers (MobileViT) as the backbone to accelerate feature extraction. Regarding the innovation of the present study, it is noticed that candidate boxes are chosen as the particles in our methodology, where the confidence is defined as the particle weight, the spatio-temopral correlation between adjacent image is used in video tracking and the Particle Filter theory is employed into the two-stage detection network to construct the present VitP-RCNN. Based on the spatial-temporal correlation, Our VitP-RCNN is efficient in predicting the particle state in the next frame through the state of the previous frame, so as to realize the continuous tracking of the landing marker. The experimental results show that on the Jetson AGX Xavier drone platform, our speed is up to 51FPS, enabling us to perform the landing marker tracking for the autonomous landing of the fixed-wing drones.INDEX TERMS Unmanned Aerial Vehicle (UAV); drone; autonomous landing; target detection; target tracking; deep learning

show abstract

Comparison of the YOLOv3 and Mask R-CNN architectures’ efficiency in the smart refrigerator’s computer vision

Cited by 15 publications

References 4 publications

Comparing YOLOv3, YOLOv4 and YOLOv5 for Autonomous Landing Spot Detection in Faulty UAVs

Comparing YOLOv3, YOLOv4 and YOLOv5 for Autonomous Landing Spot Detection in Faulty UAVs

Research on Mini-EfficientDet Identification Algorithm Based on Transfer Learning

High Speed Safe Autonomous Landing Marker Tracking of Fixed Wing Drone Based on Deep Learning

Contact Info

Product

Resources

About