In the vast, ever-changing expanse of our oceans, keeping an eye on maritime activities is a complex task. Synthetic aperture radar (SAR) has long been a go-to technology for maritime surveillance, but it’s not without its challenges. One of the main issues is that ship targets can get lost in the background noise, or ‘sea clutter’, making detection a tricky business. Enter Ran An, a researcher from the School of Information and Communication Engineering at the University of Electronic Science and Technology of China, who’s tackled this problem head-on.
An and his team have developed a novel algorithm that could significantly improve ship detection in SAR images. The algorithm, dubbed DCT-LCE (Discrete Cosine Transform-based Local Contrast Enhancement), works by transforming SAR images into the DCT domain for processing. This innovative approach uses a sliding window to characterize statistical features within the image, providing a quantitative method for distinguishing between targets and backgrounds. As An explains, “A weighted alternating current coefficients calculation method is designed to characterize statistical features within the sliding window, providing a quantitative method for distinguishing between targets and backgrounds.”
But An didn’t stop there. To further enhance the detailed morphological information of ship targets, he and his team developed an optimized version of the algorithm, called multiscale DCT local contrast enhancement (MDCT-LCE). This improvement allows for even better detection performance, both in terms of accuracy and the detailed features of the targets.
So, what does this mean for the maritime industry? Well, improved ship detection can have a significant impact on maritime safety and security. It can enhance collision avoidance systems, improve search and rescue operations, and bolster maritime surveillance and enforcement activities. Moreover, the enhanced detection capabilities can also benefit commercial shipping, offshore industries, and even environmental monitoring.
The algorithms developed by An and his team have been tested through experimental simulations, and the results are promising. Compared to other sliding window-based algorithms, DCT-LCE and MDCT-LCE show better detection performance under various levels of background complexity. These findings were recently published in the IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, a publication that translates to English as the IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.
In the words of An, “The proposed algorithms can effectively enhance ship targets.” And with these enhancements, the maritime industry can look forward to improved safety, security, and operational efficiency. It’s a significant step forward in the ongoing quest to make our oceans safer and more secure.