Objective evaluation of space multi-element remote sensing information co-detection

• Published in: 2012 International Conference on Computer Science and Information Processing (CSIP) pp. 759 - 762
• Main Authors: Zhu Weigang, Jia Xin
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.08.2012
• Subjects: formatting; insert; L-band; Measurement; style; styling
• ISBN: 1467314102; 9781467314107
• DOI: 10.1109/CSIP.2012.6308964

Space multi-element remote sensing information co-detection has been given more and more attention to these days. It urgently needs an effective objective evaluation approach to measure the effect of the co-detection. A new approach based on the geometric shape is proposed for the space multi-sensors co-detection assessment. This method extracts the extend fractal feature, and calculates the divergence of the geometric shapes between the referenced value and detection value when absent ground truth. By that means, effect of multi-sensors co-detection is measured. Experiments are conducted from two aspects to confirm the idea. Firstly, when the co-detection system includes SAR sensor, this method is more effective than the intra-region uniformity methods and the inter-region disparity methods. Secondly, when transcendental knowledge is absent, this objective evaluation appears highly consistent with other geometric shape retention evaluation metrics with the ground truth. These experiments demonstrate its general effectiveness.