Object Tracking Algorithm Based on Dual Color Feature Fusion with Dimension Reduction

Abstract

Aiming at the problem of poor robustness and the low effectiveness of target tracking in complex scenes by using single color features, an object-tracking algorithm based on dual color feature fusion via dimension reduction is proposed, according to the Correlation Filter (CF)-based tracking framework. First, Color Name (CN) feature and Color Histogram (CH) feature extraction are respectively performed on the input image, and then the template and the candidate region are correlated by the CF-based methods, and the CH response and CN response of the target region are obtained, respectively. A self-adaptive feature fusion strategy is proposed to linearly fuse the CH response and the CN response to obtain a dual color feature response with global color distribution information and main color information. Finally, the position of the target is estimated, based on the fused response map, with the maximum of the fused response map corresponding to the estimated target position. The proposed method is based on fusion in the framework of the Staple algorithm, and dimension reduction by Principal Component Analysis (PCA) on the scale; the complexity of the algorithm is reduced, and the tracking performance is further improved. Experimental results on quantitative and qualitative evaluations on challenging benchmark sequences show that the proposed algorithm has better tracking accuracy and robustness than other state-of-the-art tracking algorithms in complex scenarios.

Radar-optical fusion detection of UAV based on improved YOLOv7-tiny

“…The radar detects the target, adjusts the camera servo based on the azimuth and pitch angles of the radar-detected target, captures an image of the target using the camera, and transmits the visual information back to the data centre. Subsequently, personnel at the data centre analyse and evaluate the received information [5]. Wang et al addressed the problem of incomplete measurements in camera-tracking systems by proposing a radar-assisted cooperative attraction model and introducing a sequential fusion algorithm based on particle filtering.…”

Learning color names using least-squares probabilistic classifiers

“…Color name is one of the most important attributes used to describe images in computer vision application like image retrieval [1]- [3] object recognition [4], [5], and object tracking [6]- [8]. More specifically, for example, Schauerte and Fink [9] required color term model in human-robot interaction system for communicating a color of the intended object such as "red book", "yellow cup".…”

Visual tracking based on multi-feature fusion

“…This feature will not be affected by the translation or rotation of the image. Hu [1] et al proposed a target tracking algorithm based on two-color feature fusion, which reduces the original 11 dimensional color features to two dimensions, extracts the color features of the image, correlates the template and the candidate region through the correlation filter, obtains the color feature response of the target region, and finally estimates the target position according to the response map. Chen [2] et al proposed a multi-scale fast correlation filter tracking algorithm of feature fusion model.…”