Related Experiment Video
Updated: Jul 18, 2025
05:57
Long-term Video Tracking of Cohoused Aquatic Animals: A Case Study of the Daily Locomotor Activity of the Norway Lobster Nephrops norvegicus
Published on: April 8, 2019
6.9K
A Multiscale Method for Infrared Ship Detection Based on Morphological Reconstruction and Two-Branch Compensation
Xintao Chen1, Changzhen Qiu1, Zhiyong Zhang1
1School of Electronics and Communication Engineering, Sun Yat-sen University, Shenzhen 518107, China.
Sensors (Basel, Switzerland)
|August 26, 2023
Summary
This study introduces a new algorithm for detecting ships in infrared images, even with changing distances and complex backgrounds. The method enhances target visibility and improves accuracy for various ship sizes.
Area of Science:
- Marine surveillance
- Infrared imaging technology
- Computer vision
Background:
- Ship target detection in infrared imagery is challenging due to scale variations and complex backgrounds like sea clutter.
- Existing methods struggle with diverse environmental conditions and target sizes.
Purpose of the Study:
- To propose an effective algorithm for infrared ship target detection.
- To address challenges posed by varying ship scales and complex marine backgrounds.
- To improve detection accuracy and robustness.
Main Methods:
- Developed a multiscale morphological reconstruction method to enhance ship targets and suppress background noise.
- Incorporated a structure tensor with two feature-based filter templates to leverage contour information and boost saliency map intensities.
- Implemented a two-branch compensation strategy to handle uneven grayscale distribution in infrared images.
- Utilized an adaptive threshold for final target extraction.
Main Results:
- The proposed multiscale morphological reconstruction-based saliency mapping with a two-branch compensation strategy (MMRSM-TBC) algorithm demonstrates strong performance.
- The algorithm effectively detects ship targets across various sizes.
- Experimental results show higher accuracy compared to existing infrared ship detection methods.
Conclusions:
- The MMRSM-TBC algorithm offers a robust solution for infrared ship target detection in challenging marine environments.
- The combination of multiscale reconstruction, structure tensor analysis, and two-branch compensation significantly enhances detection capabilities.
- This approach provides a valuable advancement for marine surveillance and target recognition systems.
