Real-Time Detection, Tracking and Classification of Multiple Moving Objects in UAV Videos

Baykara, Huseyin Can; Bıyık, Erdem; Gul, Gamze; Onural, Deniz

doi:10.1109/ictai.2017.00145

Cited by 25 publications

(11 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The challenges are overcome by employing Analysis of principal components by PCANet [4] pipeline of image undistortion, image registration, classification and detections based on coordinates and velocities. Approach uses detectors like FAST, FREAK descriptors and followed by classification of Squeeze Net [5]. The workflow of candidate target generation, extracting features from candidate targets, the ground truth boxes around objects assist in tracking.…”

Section: Literature Surveymentioning

confidence: 99%

Object Detection and Tracking using Deep Learning and Artificial Intelligence for Video Surveillance Applications

Mohana¹,

Aradhya²

2019

IJACSA

View full text Add to dashboard Cite

Data is the new oil in current technological society. The impact of efficient data has changed benchmarks of performance in terms of speed and accuracy. The enhancement is visualizable because the processing of data is performed by two buzzwords in industry called Computer Vision (CV) and Artificial Intelligence (AI). Two technologies have empowered major tasks such as object detection and tracking for traffic vigilance systems. As the features in image increases demand for efficient algorithm to excavate hidden features increases. Convolution Neural Network (CNN) model is designed for urban vehicle dataset for single object detection and YOLOv3 for multiple object detection on KITTI and COCO dataset. Model performance is analyzed, evaluated and tabulated using performance metrics such as True Positive (TP), True Negative (TN), False Positive (FP), False Negative (FN), Accuracy, Precision, confusion matrix and mean Average Precession (mAP). Objects are tracked across the frames using YOLOv3 and Simple Online Real Time Tracking (SORT) on traffic surveillance video. This paper upholds the uniqueness of the state of the art networks like DarkNet. The efficient detection and tracking on urban vehicle dataset is witnessed. The algorithms give real-time, accurate, precise identifications suitable for realtime traffic applications.

show abstract

Section: Literature Surveymentioning

confidence: 99%

Object Detection and Tracking using Deep Learning and Artificial Intelligence for Video Surveillance Applications

Mohana¹,

Aradhya²

2019

IJACSA

View full text Add to dashboard Cite

show abstract

“…Nevertheless, to best of our knowledge, there is very limited literature in real-time detection and classification of human from commercial aerial vehicles. In [20], they proposed an approach to detect, track and classify humans and vehicles in real-time, using commercially available UAV systems and a common laptop computer. This study was limited to one single altitude though they presented a practical and robust method for detecting and tracking moving objects from moving videos.…”

Section: Related Workmentioning

confidence: 99%

Real Time Human Detection and Localization Using Consumer Grade Camera and Commercial UAV

Bhattarai¹,

Nakamura²,

Mozumder³

2018

Preprint

View full text Add to dashboard Cite

Human detection from Unmanned Aerial Vehicles (UAV) is gaining popularity in the field of disaster management, crowd counting, people monitoring. Real time human detection from UAV is a challenging task, because of many constraints involved. This study proposes a system for real time detection of humans on videos captured from UAVs addressing three of these constraints namely, flying height, computation time and scale of viewing. The proposed method integrated an android application with a binary classifier based on Haar-features to automatically detect human / non-human class from UAV images. The video frames were parsed and detected humans from image frames were geo-localized and visualized on Google Earth. The performance was evaluated for geo-localization accuracy, computation time and detection accuracy, considering human coverage – pixel size relationship for various heights and scale factor. Based on flying height - human size relationship and tradeoff between detection accuracy vs computation time, the study came up with optimal parameters for OpenCV’s cv2.cascadeClassifier. detectMultiScale function. This paper establishes a strong ground for further research relating to real time human detection from UAV.

show abstract

“…While missed detection still occurred. In other research by Zhang et al [15], overcoming issues from drones by performing keyframe extraction, Baykara et al [16] implemented TinyYOLO for efficient multiple object detection, and Lee et al [17] involved cloud servers by using Fast YOLO on local and Faster RCNN on a cloud. However, an error occurred during real-time testing, which implies non-effectiveness for drones in real-time conditions.…”

Section: Introductionmentioning

confidence: 99%

Exploration of The Impact of Kernel Size for YOLOv5-based Object Detection on Quadcopter

Rahmania

Corputty

Wibowo

et al. 2022

JOIV : Int. J. Inform. Visualization

View full text Add to dashboard Cite

Drones or quadcopters have been widely used in various fields based on deep learning, especially object detection. However, drone vision characteristics such as occlusion and small objects are still being explored for performance in terms of accuracy and speed detection. The YOLO architecture is very commonly used for cases requiring high-speed detection. To overcome the limitations of drone vision, in this paper, we explore the size of the YOLOv5s backbone kernel in the shallowest convolutional layer to achieve better performance. The kernel is a filter that has a main role in the feature map, and it defines the size of the convolution matrix, and the resulting features in the shallowest convolutional layer are more representative of the case of object detection and recognition. The techniques can be divided into three major categories: (1) data preprocessing, which involves augmentation and normalization of the data, (2) kernel size exploration in the shallowest convolutional layer of the YOLOv5s, and (3) model implementation in the real environment using the quadcopter. The dataset consisted of four classes representing dragon fruit, snake fruit, banana, and pineapple, with a total of 8000 data. Exploration results with kernel size give promising results. Kernel sizes 5 and 7 give an mAP of 0.988. Through these results, modification of the kernel size provides an opportunity for more in-depth investigations, such as with the epoch parameter, padding scheme, and other optimization techniques.

show abstract

Real-Time Detection, Tracking and Classification of Multiple Moving Objects in UAV Videos

Cited by 25 publications

References 16 publications

Object Detection and Tracking using Deep Learning and Artificial Intelligence for Video Surveillance Applications

Object Detection and Tracking using Deep Learning and Artificial Intelligence for Video Surveillance Applications

Real Time Human Detection and Localization Using Consumer Grade Camera and Commercial UAV

Exploration of The Impact of Kernel Size for YOLOv5-based Object Detection on Quadcopter

Contact Info

Product

Resources

About