摘要
基于改进遗传算法提出了一种新型的自适应阈值。由于实际检测过程中受光照强度以及气候等外部因素的影响,当目标图像的配置阈值与实际阈值相差较大时,检测结果往往不够准确。针对不同的图像,通过遗传算法计算出最优阈值,并将计算出的阈值用于边缘检测。引入图像评价指标对本算法在处理尾流图像的效果进行定量评估,结果表明,与传统的边缘检测算法相比,本文所改进的算法处理后的尾流图像边缘连续性更强,在尾流图像处理中具有更好的检测效果和适用性。对比结果与主观对比检测结果图像得出的结论相同,说明本算法可以更准确提取尾流的边缘,且失真较小。
A novel adaptive threshold Sobel edge detection algorithm based on an improved genetic algorithm is proposed in this paper to improve the edge detection and extraction of effective information in wake images.Due to the influence of external factors in the actual detection process,when the configured threshold of the target image is much different from the actual threshold,the detection result is often not accurate.For different images,optimal thresholds are calculated using a genetic algorithm and these thresholds are then utilized for edge detection.To quantitatively assess the algorithm’s performance in processing wake images,image evaluation metrics are employed.The results demonstrate that,when compared to traditional edge detection algorithms,the improved algorithm proposed in this paper shows improved edge continuity,superior detection performance,and broader applicability in wake image processing.The comparison outcomes are consistent with the conclusions derived from subjective detection comparisons,indicating that this algorithm can extract wake edges with higher accuracy and reduced distortion.
作者
姚远
张建生
闫林波
焦贵金
董敏
YAO Yuan;ZHANG Jiansheng;YAN Linbo;JIAO Guijin;DONG Min(School of Sciences,Xi’an Technological University,Xi’an 710021,China)
出处
《西安工业大学学报》
2025年第4期683-690,共8页
Journal of Xi’an Technological University
基金
陕西省重点研发计划项目(2023-YBGY-016)。
关键词
边缘检测
尾流
遗传算法
图像处理
边缘评价
edge detection
wake
genetic algorithm
image processing
edge evaluation
