Research on fusion technology based on low-light visible image and infrared image

Abstract

Abstract. Image fusion technology usually combines information from multiple images of the same scene into a single image so that the fused image is often more informative than any source image. Considering the characteristics of low-light visible images, this study presents an image fusion technology to improve contrast of lowlight images. This study proposes an adaptive threshold-based fusion rule. Threshold is related to the brightness distribution of original images. Then, the fusion of low-frequency coefficients is determined by threshold. Pulsecoupled neural networks (PCNN)-based fusion rule is proposed for fusion of high-frequency coefficients. Firing times of PCNN reflect the amount of detail information. Thus, a high-frequency coefficient corresponding to maximum firing times is chosen as the fused coefficient. Experimental results demonstrate that the proposed method obtains high-contrast images and outperforms traditional fusion approaches on image quality. © The Authors. Published by SPIE under a Creative Commons Attribution 3.0 Unported License. Distribution or reproduction of this work in whole or in part requires full attribution of the original publication, including its DOI.

Airborne Infrared and Visible Image Fusion Combined with Region Segmentation

“…The comparison methods include: classic multiresolution decomposition image fusion method DWT [12]; the PCA-MST method proposed by Li et al [22], which bases image multiresolution decomposition on Contourlet transform domain and determines the low-frequency fusion coefficient through PCA and the high-frequency fusion coefficient through the average gradient method; the HIS-PCNN method proposed by Liu et al [23], which enhances the image contrast through IHS transform and fuses low-frequency components through the adaptive threshold rule and high-frequency components through PCNN; and the GFF method proposed by Li et al [24], which breaks down the source image into a basic layer and a detail layer and then fuses the two layers by using the GFF-based weighted average method.…”

“…In contrast, in the visible light image for the same scene, the pedestrians in the dark can't be seen clearly, but the billboard information completely invisible in the IR image is legible here. Then the UNcamp images are fused by using the proposed method and other four methods including DWT [12], PCA-MST [22], IHS-PCNN [23] and GFF [24]. The results of fusion comparisons are shown in the Figure 6.…”

“…Quad image fusion comparisons: ( a ) IR image; ( b ) Visible image; ( c ) discrete wavelet transform (DWT) [12]; ( d ) PCA-MST [22]; ( e ) guidance filter (GFF) [24]; ( f ) IHS-PCNN [23]; ( g ) The proposed method.…”

“…In the image fused by this method, the information on hot targets is relatively complete, but the information on background details is blurred. To better keep the information on background details of the visible image, Liu et al [23] proposed the image fusion method combining PCNN (pulse-coupled neural networks) with IHS transform. In this, the image contrast was enhanced through IHS transform, low-frequency components were fused through the rule based on adaptive threshold, and high-frequency components were fused through PCNN.…”

Infrared and Visible Image Fusion based on Saliency Detection and Infrared Target Segment

“…On the contrary, a visible image is usually suitable for the human visual system. It has refined sharpness and high spatial resolution, which can provide the geometry and texture detail information of the scene [3][4][5][6]. It is apparent that the _________________________________________ infrared image and visible image offer complementary advantages.…”