Activity Feed

13096300 1739689082982412 1116910342765371095 n

Shuo L. added photos to project Multi-Channel CNN-based Object Detection for Enhanced Situation Awareness

Medium 2d98e07c 9a84 435b 99ce e4fb3920c4fc

Multi-Channel CNN-based Object Detection for Enhanced Situation Awareness

Object Detection is critical for automatic military operations. However, the performance of current object detection algorithms is deficient in terms of the requirements in military scenarios. This is mainly because the object presence is hard to detect due to the indistinguishable appearance and dramatic changes of object's size which is determined by the distance to the detection sensors. Recent advances in deep learning have achieved promising results in many challenging tasks. The state-of-the-art in object detection is represented by convolutional neural networks (CNNs), such as the fast R-CNN algorithm. These CNN-based methods improve the detection performance significantly on several public generic object detection datasets. However, their performance on detecting small objects or undistinguishable objects in visible spectrum images is still insufficient. In this study, we propose a novel detection algorithm for military objects by fusing multiple images using CNNs. We combine spatial, temporal and thermal information by generating a three-channel image, and they will be fused as CNN feature maps in an unsupervised manner. The backbone of our object detection framework is from the fast R-CNN algorithm, and we utilize cross-domain transfer learning technique to fine-tune the CNN model on generated multi-channel images. In the experiments, we validated the proposed method with the images from SENSIAC (Military Sensing Information Analysis Centre) database and compared it with the state-of-the-art. The experimental results demonstrated the effectiveness of the proposed method on both accuracy and computational efficiency.

13096300 1739689082982412 1116910342765371095 n

Shuo L. added photos to project Multi-Channel CNN-based Object Detection for Enhanced Situation Awareness

Medium 0c371bc5 65b4 4d19 a3e0 ddaabc1da5a6

Multi-Channel CNN-based Object Detection for Enhanced Situation Awareness

Object Detection is critical for automatic military operations. However, the performance of current object detection algorithms is deficient in terms of the requirements in military scenarios. This is mainly because the object presence is hard to detect due to the indistinguishable appearance and dramatic changes of object's size which is determined by the distance to the detection sensors. Recent advances in deep learning have achieved promising results in many challenging tasks. The state-of-the-art in object detection is represented by convolutional neural networks (CNNs), such as the fast R-CNN algorithm. These CNN-based methods improve the detection performance significantly on several public generic object detection datasets. However, their performance on detecting small objects or undistinguishable objects in visible spectrum images is still insufficient. In this study, we propose a novel detection algorithm for military objects by fusing multiple images using CNNs. We combine spatial, temporal and thermal information by generating a three-channel image, and they will be fused as CNN feature maps in an unsupervised manner. The backbone of our object detection framework is from the fast R-CNN algorithm, and we utilize cross-domain transfer learning technique to fine-tune the CNN model on generated multi-channel images. In the experiments, we validated the proposed method with the images from SENSIAC (Military Sensing Information Analysis Centre) database and compared it with the state-of-the-art. The experimental results demonstrated the effectiveness of the proposed method on both accuracy and computational efficiency.

13096300 1739689082982412 1116910342765371095 n

Shuo L. added photos to project Multi-Channel CNN-based Object Detection for Enhanced Situation Awareness

Medium 85896a5d 7f7e 480e 9efa 9c74a495c809

Multi-Channel CNN-based Object Detection for Enhanced Situation Awareness

Object Detection is critical for automatic military operations. However, the performance of current object detection algorithms is deficient in terms of the requirements in military scenarios. This is mainly because the object presence is hard to detect due to the indistinguishable appearance and dramatic changes of object's size which is determined by the distance to the detection sensors. Recent advances in deep learning have achieved promising results in many challenging tasks. The state-of-the-art in object detection is represented by convolutional neural networks (CNNs), such as the fast R-CNN algorithm. These CNN-based methods improve the detection performance significantly on several public generic object detection datasets. However, their performance on detecting small objects or undistinguishable objects in visible spectrum images is still insufficient. In this study, we propose a novel detection algorithm for military objects by fusing multiple images using CNNs. We combine spatial, temporal and thermal information by generating a three-channel image, and they will be fused as CNN feature maps in an unsupervised manner. The backbone of our object detection framework is from the fast R-CNN algorithm, and we utilize cross-domain transfer learning technique to fine-tune the CNN model on generated multi-channel images. In the experiments, we validated the proposed method with the images from SENSIAC (Military Sensing Information Analysis Centre) database and compared it with the state-of-the-art. The experimental results demonstrated the effectiveness of the proposed method on both accuracy and computational efficiency.

Medium 13096300 1739689082982412 1116910342765371095 n

Shuo L. created project Multi-Channel CNN-based Object Detection for Enhanced Situation Awareness

Medium 85896a5d 7f7e 480e 9efa 9c74a495c809

Multi-Channel CNN-based Object Detection for Enhanced Situation Awareness

Object Detection is critical for automatic military operations. However, the performance of current object detection algorithms is deficient in terms of the requirements in military scenarios. This is mainly because the object presence is hard to detect due to the indistinguishable appearance and dramatic changes of object's size which is determined by the distance to the detection sensors. Recent advances in deep learning have achieved promising results in many challenging tasks. The state-of-the-art in object detection is represented by convolutional neural networks (CNNs), such as the fast R-CNN algorithm. These CNN-based methods improve the detection performance significantly on several public generic object detection datasets. However, their performance on detecting small objects or undistinguishable objects in visible spectrum images is still insufficient. In this study, we propose a novel detection algorithm for military objects by fusing multiple images using CNNs. We combine spatial, temporal and thermal information by generating a three-channel image, and they will be fused as CNN feature maps in an unsupervised manner. The backbone of our object detection framework is from the fast R-CNN algorithm, and we utilize cross-domain transfer learning technique to fine-tune the CNN model on generated multi-channel images. In the experiments, we validated the proposed method with the images from SENSIAC (Military Sensing Information Analysis Centre) database and compared it with the state-of-the-art. The experimental results demonstrated the effectiveness of the proposed method on both accuracy and computational efficiency.

Medium 13096300 1739689082982412 1116910342765371095 n

Shuo L. updated status

Medium 13096300 1739689082982412 1116910342765371095 n

Shuo Liu

Visual automatic surveillance is an attractive research area in image processing, computer vision, and machine learning, due to its essential practical applications in the real world. And my research topic is on its crucial component, visual object detection, which is an algorithm that has the ability to automatically localize and recognize the visual objects in an image or a video. This research topic is also the key to several advanced unmanned systems, such as the vehicle detection in the autonomous driving system. However, it still has much room need to be improved for practical applications. In my research, I'm working on designing a visual object detection algorithm which adopts deep learning technologies and image fusion techniques to advance state-of-the-art object detection system.

About

A graduate student at the University of British Columbia, Okanagan. My research interests include image processing, computer vision, deep learning, pattern recognition and information fusion.

Featured Projects

See All

No groups to show at the moment.

Medium real sense
Featured
  • Followers 1727

Intel RealSense™

Natural interaction, immersive, collaboration, gaming and learning, 3D scanning. Integrate tracki...

Medium android
Featured
  • Followers 1749

Android

Intel is inside more and more Android devices, and we have tools and resources to make your app d...

Medium big data
Featured
  • Followers 1813

Modern Code

Drive faster breakthroughs through faster code: Get more results on your hardware today and carry...

Medium networking
Featured
  • Followers 1661

Networking

Software-Defined Networking (SDN) and Network Functions Virtualization (NFV) are transforming the...

Medium achievement unlocked logo
Featured
  • Followers 1584

Game Development

Upgrade your skills as a game developer, share your game projects, and connect with other develop...

Medium vr large 575x441
Featured
  • Followers 1295

Virtual Reality

VR, AR, mixed reality...you'll find projects based on all these new platforms here. Share your own!

No users to show at the moment.