Infrared and Visible Light Images Fusion Utilizing Object Detection

Abstract

针对红外图像和可见光图像融合，提出了一种基于目标检测的图像融合方法。该方法运用目标检测技术将红外图像分割为目标和背景区域，同时运用非下采样轮廓波变换将红外背景区域和可见光图像融合得到融合背景图像，然后将融合背景图像和红外目标区域融合得到最终融合图像。实验结果表明，该方法具有可行性和高效性，且比其他图像融合方法具有更好的性能。 A new image fusion method for infrared and visible light image based on object detection technique is proposed in this paper. The object detection is utilized to separate infrared image into object and background region. And NSCT is applied to fuse the background and visible light image. The final fused image is obtained by fusing the background fused image and infrared object image. Fusion experiments indicate that the proposed method is efficient and effective. Besides, it achieves better performance than the conventional fusion methods.