Acta Optica Sinica, Volume. 30, Issue s1, 100408(2010)
Image Mosaicking Quality Objective Assessment Method Based on Visual Information Fidelity
Objective quality assessment of image mosaic is of fundamental necessary in real-time digital image mosaic system. An image mosaicking quality objective assessment method based on visual information fidelity(VIF) is proposed considering that the results using traditional objective assessment methods do not agree with the results of the human subjective assessment methods. Firstly, assuming that the image sources meet Gaussian scale mixture (GSM) model and regarding image mosaic algorithms as image signal distortion channel, and considering human visual noise characteristics within neurons, VIF digital image mosaic objective quality assessment model is established, and a mosaic image quality evaluation index mosaic visual information fidelity (MVIF) is also derived. And then, the method of estimating the parameters in MVIF assessment algorithm is introduced in detail. Finally, the performance of MVIF assessment algorithm has been validated with lots of mosaic images. The experiment indicates that the method can evaluate mosaic image quality correctly and its objective evaluation results are more consistent with subjective evaluation results than those of traditional methods.
Get Citation
Copy Citation Text
Yu Hongsheng, Jin Weiqi. Image Mosaicking Quality Objective Assessment Method Based on Visual Information Fidelity[J]. Acta Optica Sinica, 2010, 30(s1): 100408
Category: Fourier optics and signal processing
Received: Jun. 1, 2010
Accepted: --
Published Online: Dec. 17, 2010
The Author Email: Hongsheng Yu (yhs99981@sina.com)