A shadow-removal based saliency map for point feature detection of underwater objects

2015 ◽  
Author(s):  
Liqin Fu ◽  
Yiru Wang ◽  
Zhebin Zhang ◽  
Rui Nian ◽  
Tianhong Yan ◽  
...  
Keyword(s):  
Feature Detection ◽  
Saliency Map ◽  
Shadow Removal ◽  
Point Feature

Multi-Frame Image Fusion Method Combining Spatial-Temporal Saliency Detection and NSCT

2011 ◽  
Vol 403-408 ◽  
pp. 1927-1932
Author(s):  
Hai Peng ◽  
Hua Jun Feng ◽  
Ju Feng Zhao ◽  
Zhi Hai Xu ◽  
Qi Li ◽  
...  
Keyword(s):  
Image Fusion ◽  
Feature Detection ◽  
Saliency Detection ◽  
Detection Algorithm ◽  
Saliency Map ◽  
Image Sequences ◽  
Fusion Method ◽  
Threshold Method ◽  
Salient Features ◽  
Image Fusion Method

We propose a new image fusion method to fuse the frames of infrared and visual image sequences more effectively. In our method, we introduce an improved salient feature detection algorithm to achieve the saliency map of the original frames. This improved method can detect not only spatially but also temporally salient features using dynamic information of inter-frames. Images are then segmented into target regions and background regions based on saliency distribution. We formulate fusion rules for different regions using a double threshold method and finally fuse the image frames in NSCT multi-scale domain. Comparison of different methods shows that our result is a more effective one to stress salient features of target regions and maintain details of background regions from the original image sequences.

Fully automatic hybrid registration method based on point feature detection without user intervention

2006 ◽  
Author(s):  
Bang-Bon Koo ◽  
Jong-Min Lee ◽  
June-Sic Kim ◽  
In-Young Kim ◽  
Jun-Soo Kwon ◽  
...  
Keyword(s):  
Feature Detection ◽  
Registration Method ◽  
Fully Automatic ◽  
User Intervention ◽  
Hybrid Registration ◽  
Point Feature

A deep neural network-based vehicle re-identification method for bridge load monitoring

2021 ◽  
pp. 136943322110339
Author(s):  
Yufeng Zhang ◽  
Junxin Xie ◽  
Jiayi Peng ◽  
Hui Li ◽  
Yong Huang
Keyword(s):  
Neural Network ◽  
Feature Detection ◽  
Image Data ◽  
Image Features ◽  
Superior Performance ◽  
Multiple Sources ◽  
Identification Method ◽  
Load Monitoring ◽  
Bridge Structures ◽  
Point Feature

The accurate tracking of vehicle loads is essential for the condition assessment of bridge structures. In recent years, a computer vision method that is based on multiple sources of data from monitoring cameras and weight-in-motion (WIM) systems has become a promising strategy in bridge vehicle load identification for structural health monitoring (SHM) and has attracted increasing attention. The implementation of vehicle re-identification, namely, the identification of the same vehicle from images that were captured at different locations or time instants, is the key topic of this study. In this study, a vehicle re-identification method that is based on HardNet, a deep convolutional neural network (CNN) specialized in picking up local image features, is proposed. First, we obtain the vehicle point feature positions in the image through feature detection. Then, the HardNet is employed to encode the point feature image patches into deep learning feature descriptors. Re-identification of the target vehicle is achieved by matching the encoded descriptors between two images, which are robust toward scaling, rotation, and other types of noises. A comparison study of the proposed method with three published vehicle re-identification methods is performed using vehicle image data from a real bridge, and the superior performance of our proposed method is demonstrated.

scholarly journals A generalised framework for saliency-based point feature detection

2017 ◽  
Vol 157 ◽  
pp. 117-137 ◽  
Author(s):  
Mark Brown ◽  
David Windridge ◽  
Jean-Yves Guillemaut
Keyword(s):  
Feature Detection ◽  
Point Feature

Shadow detection and correction using a combined 3D GIS and image processing approach

2019 ◽  
Vol 29 (3-4) ◽  
pp. 241-253
Author(s):  
Safa Ridene ◽  
Reda Yaagoubi ◽  
Imane Sebari ◽  
Audrey Alajouanine
Keyword(s):  
Image Analysis ◽  
Visual Processing ◽  
Feature Detection ◽  
Poor Quality ◽  
Shadow Removal ◽  
Quality Of Information ◽  
Multiple Image ◽  
Shadow Cast ◽  
Objects Detection

While shadow can give useful information about size and shape of objects, it can pose problems in feature detection and object detection, thereby, it represents one of the major perturbator phenomenons frequently occurring on images and unfortunately, it is inevitable. “Shadows may lead to the failure of image analysis processes and also cause a poor quality of information which in turn leads to problems in implementation of algorithms.” (Mahajan and Bajpayee, 2015). It also affects multiple image analysis applications, whereby shadow cast by buildings deteriorate the spectral values of the surfaces. Therefore, its presence causes a deterioration in the visual image's quality and limits the information that the former could give. Ignoring the existence of shadows in images may cause serious problems in various visual processing applications such as false objects detection. In this context, many researches have been conducted through years. However, it is still a challenge for analysts all over the world to find a fully automated and efficient method for shadow removal from images.

Feature Detection Requires Spatial Attention

2007 ◽  
Author(s):  
Jan Theeuwes ◽  
Erik van der Burg ◽  
Artem V. Belopolsky
Keyword(s):  
Spatial Attention ◽  
Feature Detection
Sign in / Sign up

Export Citation Format

Share Document