A new method to retrieve rainfall intensity level from rain-contaminated X-band marine radar image

ABSTRACT

In this study, a new method combining wave-number energy spectrum (WES) and Genetic algorithm-back propagation neural network (GA-BPNN) is proposed to retrieve the rainfall intensity level from rain-contaminated X-band marine radar image. Since the intensity of spatial rainfall can be reflected by the distribution of energy in the wavenumber frequency domain, the obtained WES is divided into three wavenumber segments (low, medium and high wavenumber segments), and the ratio of the wavenumber in each wavenumber segment to the total wavenumber is calculated separately as the characteristic parameters. Based on the excellent network convergence speed and data prediction accuracy of the GA-BPNN, these calculated parameters are input into the constructed GA-BPNN for training to complete the task of rainfall intensity level retrieval. The proposed method is tested using data collected at the ocean observation station of Haitan Island in Pingtan County. Referring to the actual rainfall intensity synchronously recorded by the rain gauge, the retrieval accuracy of the proposed method is 97.4%, which is 4.3% higher than that of back propagation neural network (BPNN) not optimized by Genetic algorithm (GA). In addition, compared with the retrieval performance of the ratio of zero intensity to echo (RZE) method based on the occlusion area of radar image, the retrieval accuracy of the proposed method is improved by about 12.9%.

KEYWORDS:

• Wave-number energy spectrum (WES)
• genetic algorithm-back propagation neural network (GA-BPNN)
• rainfall intensity level retrieval
• radar image

Acknowledgements

The authors greatly appreciate the editors and anonymous reviewers for their efforts and time which they have spent on this article.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by the Natural Science Foundation of Jiangsu Province of China under Grant (No. BK20180988) and National Natural Science Foundation of China under Grant (No. 41906154)