Vehicle Detection in Aerial Images Based on 3D Depth Maps and Deep Neural Networks

Javadi, Saleh; Dahl, Mattias; Pettersson, Mats I.

doi:10.1109/access.2021.3049741

Cited by 31 publications

(14 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Based on the route with less traffic volume, less density, less distance to the destination, and high vehicle speed, the optimal path is selected. The fitness of the solution is computed by using Equation (18). The objective is to minimize the fitness.…”

Section: Fitness Evaluationmentioning

confidence: 99%

“…The miss in detection is prevented with the suggested sensor setup as there is less chance of the sensor being by passed. Javadi et al 18 presented a unique method for detecting vehicles from aerial images using deep neural networks and 3D feature maps. The vehicles were detected using a YOLOv3 detector that includes DenseNet‐201, MobileNet‐v2, Squeeze Net, and DarkNet‐53.…”

Section: Literature Reviewmentioning

confidence: 99%

See 1 more Smart Citation

Toward a deep CNN and RS‐GOA framework for vehicle detection, traffic flow estimation, and optimal path selection from surveillance videos

Sasikala

Neelaveni

Jose

2023

Int J Communication

View full text Add to dashboard Cite

SummarySeveral autonomous traffic monitoring systems have been created as a result of the growing number of vehicles in urban areas. Traffic surveillance systems that use roadside cameras, in particular, are becoming widely used for traffic management. For an efficient traffic control and vehicle navigation system, accurate traffic flow information must be obtained based on the vehicles detected in surveillance videos. However, vehicles of various scales are difficult to spot in traffic surveillance videos due to the presence of barricades, other vehicles, and the impact of poor lighting. Also, adverse weather conditions like snow, fog, and heavy rain diminish the visual quality of the surveillance footage. This paper proposes multi‐scale dense nested deep CNN (MSDN‐DCNN) and regional search grasshopper optimization algorithm (RS‐GOA) framework to accurately detect the vehicles, estimate the traffic flow, and find the optimal path with less travel time. First, the surveillance videos are pre‐processed, which includes frame conversion, redundancy removal, and image enhancement. The pre‐processed frames are given as input to the MSDN‐DCNN for multi‐scale vehicle detection. The detected results are used for vehicle counting and estimating the traffic flow. Finally, the optimal path is chosen based on the traffic flow information by using the RS‐GOA algorithm. The performance of the proposed method is compared with the existing vehicle detection and path selection techniques. The results illustrate that the proposed Deep CNN‐RS‐GOA framework has improved performance with high detection accuracy (91.03%), high speed (53.9 fps), less running time (1,000 ms), less travel time, and faster convergence.

show abstract

Section: Fitness Evaluationmentioning

confidence: 99%

Section: Literature Reviewmentioning

confidence: 99%

Toward a deep CNN and RS‐GOA framework for vehicle detection, traffic flow estimation, and optimal path selection from surveillance videos

Sasikala

Neelaveni

Jose

2023

Int J Communication

View full text Add to dashboard Cite

show abstract

“…[79] Comparison among faster R-CNN, R-FCN, and SSD (Best model) [80] Optimized DL model considering feature extraction, object detection, and non-maximum suppression. [81] Small-Sized Vehicle Detection Network (AVDNet) (one-stage vehicle detection network) [82] Comparison among four object detection networks: D-YOLO (best model), YOLOV2, YOLOV3, and YOLT [83] Vehicle detection based on RetinaNet architecture [84] Model based on Alexnet network (classification) and Faster R-CNN (target detection) [85] Faster R-CNN with a improved feature-balanced pyramid network (FBPN) [86] Comparison among YOLOv3, YOLOv4 (best models), and Faster R-CNN [87] Super-resolution cyclic GAN with RFA and YOLO as the detection network (SRCGAN-RFA-YOLO) 1, 2 [88] Modified YOLOv3 and fcNN using 3D features in cascade. [16] Method using the lightweight feature extraction network with the Faster R-CNN [17] Orientation-Aware Vehicle Detection with an Anchor-Free Object Detection approach…”

Section: Papermentioning

confidence: 99%

Bounding Box-Free Instance Segmentation Using Semi-Supervised Iterative Learning for Vehicle Detection

Carvalho

Albuquerque

Santana

et al. 2022

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

Vehicle classification is a hot computer vision topic, with studies ranging from ground-view to top-view imagery. Topview images allow understanding city patterns, traffic management, among others. However, there are some difficulties for pixel-wise classification: (a) most vehicle classification studies use object detection methods, and most publicly available datasets are designed for this task, (b) creating instance segmentation datasets is laborious, and (c) traditional instance segmentation methods underperform on this task since the objects are small. Thus, the present research objectives are: (1) propose a novel semi-supervised iterative learning approach using GIS software, (2) propose a box-free instance segmentation approach, and (3) provide a city-scale vehicle dataset. The iterative learning procedure considered: (1) labeling a few vehicles from the entire scene, (2) choosing training samples near those areas, ( 3) training the DL model (U-net with Efficient-net-B7 backbone), (4) classifying the whole scene, (5) converting the predictions into shapefile, (6) correcting areas with wrong predictions, (7) including them in the training data, (8) repeating until results are satisfactory. We considered vehicle interior and borders to separate instances using a semantic segmentation model. When removing the borders, the vehicle interior becomes isolated, allowing for unique object identification. Our procedure is very efficient and accurate for generating data iteratively, which resulted in 122,567 mapped vehicles. Metrics-wise, our method presented higher IoU when compared to box-based methods (82% against 72%), and per-object metrics surpassed 90% for precision and recall.

show abstract

“…The differences between conventional nature images and drone images make OD di cult. First, objects in such images are scaled differently [7]. Far objects are small, while near ones are big.…”

Section: Introductionmentioning

confidence: 99%

An enhanced YOLOv5 based on Color Harmony Algorithm for object detection in Unmanned Aerial Vehicle captured Images

Singh

Jain

2022

Preprint

View full text Add to dashboard Cite

Recent improvements in robotics and computer vision enable new camera-equipped drone applications. Aerial Object Detection (OD) is one. Despite recent advances, computer vision OD remains difficult. Due to UAVs' fast speed, different views, and fluctuating altitudes, objects in Unmanned Aerial Vehicle (UAV) photos are heterogeneous, fluctuate in size, and are dense, making OD challenging with existing algorithms. Existing object recognition algorithms perform worse on UAV images because OD in aerial images is more difficult than in ground-taken images. If a generic object detection technique is used to drone-captured images, its performance will be drastically degraded owing to the fact that varied surroundings with complicated background and the size of objects are at the core of this phenomena. In this work, an object detection model CHA-YOLOv5 is proposed. In the proposed system the detection module is optimized using the Color Harmony Algorithm, as it determines the prediction head from the available three prediction heads. CHA-YOLOv5 accurately predicts several bounding boxes per grid cell. The proposed model has been trained using several images from challenging contexts. The experimental results show that the objects are identified accurately using the proposed improved network model in this study. It achieved precision 98.675%, pecall 97.8023%, mAP 97.23%, F1 score 98.50%. Our system outperformed the YOLOV5, ResNet50, VGG19, and InceptionNet V3.

show abstract

Vehicle Detection in Aerial Images Based on 3D Depth Maps and Deep Neural Networks

Cited by 31 publications

References 40 publications

Toward a deep CNN and RS‐GOA framework for vehicle detection, traffic flow estimation, and optimal path selection from surveillance videos

Toward a deep CNN and RS‐GOA framework for vehicle detection, traffic flow estimation, and optimal path selection from surveillance videos

Bounding Box-Free Instance Segmentation Using Semi-Supervised Iterative Learning for Vehicle Detection

An enhanced YOLOv5 based on Color Harmony Algorithm for object detection in Unmanned Aerial Vehicle captured Images

Contact Info

Product

Resources

About