Characterisation of image fusion quality metrics for surveillance applications over bandlimited channels

Abstract

-Image fusion isfinding increasing application in areas such as medical imaging, remote sensing or military surveillance using sensor networks. Many ofthese applications demand highly compressed data combined with error resilient coding due to the characteristics of the communication channeL In this respect, JPEG2000 has many advantages over previous image coding standards.This paper evaluates and compares quality metricsfor lossy compression using JPEG2000. Three representative image fusion algorithms: simple averaging, contrast pyramid and dual-tree complex wavelet transform basedfusion have been considered. Numerous infrared and visible test images have been used. We compare these results with a psychophysical study where participants were asked to perform specific tasks and assess imagefusion quality.The results show that there is a correlation between most of the metrics and the psychophysical evaluation. They also indicate that selection of the correct fusion method has more impact on performance than the presence ofcompression.

Scanpath Analysis of Fused Multi-Sensor Images with Luminance Change: A Pilot Study

“…These advantages come at the cost of greater computational expense. This fusion scheme has been shown to produce better results than other DWT methods [4,5], as well as other pyramid and averaging methods [6,7], across a range of qualitative and quantitative assessments. In the current paper, a simple averaging scheme (AVE) was also used, for reference.…”

“…Initiated by Toet and colleagues [8,9], definite advances have been made in applying some form of task to the assessment process, and moving away from the ever-present subjective quality assessment. Furthermore, in recent findings [5][6][7], it has been shown that objective task results can differ significantly from subjective ratings. It is thus essential to choose a good task when assessing fused images or video sequences, and to go beyond simply applying a subjective rating to the fused outputs.…”

“…The problem often found with basic metrics is that in the past they have been found not to correlate well with subjective ratings [1,10]. As has been shown in work carried by Dixon and colleagues [5][6][7], metric comparison with objective human task results might overcome this explanatory gap. More recent metrics based on aspects of the human visual system might be able to overcome some of these shortcomings [1].…”

Image Fusion in the JPEG 2000 Domain

“…These images were then, either fused in the JPEG 2000 domain or decompressed and then fused with other widely used methods (CP, DT-CWT or DWT). Quality values for the fused images were computed using Piella's metric: the Image Fusion Quality Index (IFQI) [10], and Petrovic's metric [11] (see [1] for a review on these metrics).…”

“…Due to the characteristics of the communication channel, many of these applications demand highly compressed data combined with error resilient coding. In this respect, JPEG 2000 has many advantages over previous image coding standards [1], [2], [3], [4], [5].…”