Face image quality improvement algorithm based on illumination adaptive adjustment and fuzzy classification
-
摘要: 针对铁路人脸识别闸机使用中影响人脸识别准确率的光照和模糊问题,文章提出一种人脸光照自适应调节算法,以提升非理想光照条件下的人脸识别准确率;设计了模糊识别模块,以挑选清晰的人脸图像,提升旅客移动场景的人脸识别准确率。在自建数据集中进行算法实验,实验结果表明,采用文章设计的算法,人脸识别准确率达到97.21%,能够满足实际应用的需求,具有推广价值。Abstract: Aiming at the illumination and blurring problems that affect the accuracy of face recognition in the use of railway face recognition gates, this paper proposed a face illumination adaptive adjustment algorithm to improve the accuracy of face recognition under non ideal lighting conditions, designed a clear fuzzy identification algorithm for face images to select clear face images and improve the accuracy of face recognition in passenger moving scenes. The algorithm experiments were carried out in self-built data sets. Experimental results show that the accuracy rate of face recognition reaches 97.21% using the algorithm designed in this paper, which can meet the needs of practical applications and has the value of popularization..
-
-
表 1 不同模型的性能参数及准确率
模型名称 其他变量 阈值 准确率 MobileNetV2 / 0.58 0.9684 模糊识别模型-A / 0.56 0.9691 模糊识别模型-B 局部赋权模块 0.32 0.9701 模糊识别模型-C 局部赋权模块+Arcface损失函数 0.3 0.9721 -
[1] 贾成强,戴琳琳,徐海涛,等. 基于人脸识别技术的铁路实名制进站核验系统研究及设计 [J]. 铁路计算机应用,2018,27(7):49-53,63. DOI: 10.3969/j.issn.1005-8451.2018.07.012 [2] 景 辉,阎志远,戴琳琳,等. 基于Faster R-CNN的人脸识别算法研究 [J]. 铁路计算机应用,2019,28(10):8-11. DOI: 10.3969/j.issn.1005-8451.2019.10.003 [3] 戴琳琳,阎志远,景 辉. Insightface结合Faiss的高并发人脸识别技术研究 [J]. 铁路计算机应用,2020,29(10):16-20. DOI: 10.3969/j.issn.1005-8451.2020.10.004 [4] 王 萍. 基于局部失焦模糊的拼接图像盲检测研究 [J]. 软件工程,2017,20(2):16-18. DOI: 10.3969/j.issn.1008-0775.2017.02.005 [5] 董星煜,刘传奇,赵健康. 融合运动信息的图像运动模糊去除算法 [J]. 计算机应用研究,2021,38(1):278-281. [6] 王 晶,苏光大. 基于方向滤波和自商图像的人脸光照补偿 [J]. 光子学报,2010,39(9):1641-1644. [7] 金玉多. 改进Gamma校正的人脸图像光照预处理算法 [J]. 电脑知识与技术,2019,15(20):214-217. [8] 刘丽华,王映辉,邓方安. 基于小波商图像的人脸光照补偿 [J]. 计算机工程与设计,2009,30(14):3402-3405. [9] 刘丽华. 基于线性子空间和商图像理论的人脸光照补偿 [J]. 计算机工程与应用,2008,44(25):189-192,218. DOI: 10.3778/j.issn.1002-8331.2008.25.057 [10] 王怀斌,王海涛,高凌飞,等. 基于人脸识别的人脸图像质量评估 [J]. 计算技术与自动化,2021,40(3):122-126. [11] Terhörst P, Kolf J N, Damer N, et al. SER-FIQ: unsupervised estimation of face image quality based on stochastic embedding robustness[C]//2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition, 13-19 June, 2020, Seattle, WA, USA. New York, USA: IEEE, 2020: 5650-5659.
[12] 胡正平,潘佩云,郑赛月,等. 非对称内核卷积结合语义置信嵌入的模糊人脸图像重建 [J]. 模式识别与人工智能,2021,34(7):646-654. [13] Wang X T, Li Y, Zhang H L, et al. Towards real-world blind face restoration with generative facial prior[C]//2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition, 20-25 June, 2021, Nashville, TN, USA. New York, USA: IEEE, 2021: 9168-9178.
[14] 唐 述,万盛道,杨书丽,等. 空间尺度信息的运动模糊核估计方法 [J]. 软件学报,2019,30(12):3876-3891. DOI: 10.13328/j.cnki.jos.005587 [15] 何 涛,赵 停,徐 鹤. 基于暗通道先验的单幅图像去雾新算法 [J]. 计算机科学,2021,48(7):219-224. DOI: 10.11896/jsjkx.200700160 [16] 许长路,李林升,立济伟. 基于Gamma矫正与LOG算法融合的锂电池极片缺陷检测方法 [J]. 机械设计与研究,2020,36(6):105-109. [17] Sandler M, Howard A, Zhu M L, et al. MobileNetV2: inverted residuals and linear bottlenecks[C]//Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, 18-23 June, 2018, Salt Lake City, UT, USA. New York, USA: IEEE, 2018: 4510-4520.
[18] Deng J K, Guo J, Xue N N, et al. ArcFace: additive angular margin loss for deep face recognition[C]//Proceedings of the 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition, 15-20 June, 2019, Long Beach, CA, USA. New York, USA: IEEE, 2019: 4690-4699.
-
期刊类型引用(3)
1. 孔垂云. 基于货运票据电子化的货物全程追踪与预计到达关键技术. 中国铁路. 2021(11): 14-18 . 百度学术
2. 李进良,高俊杰. 基于北斗卫星与惯性传感器组合导航技术的现代有轨电车定位终端设计. 铁路计算机应用. 2019(10): 60-62 . 本站查看
3. 郄晓永. 嵌入式无线货车装载状态高清检测系统方案研究. 铁路计算机应用. 2018(12): 13-17 . 本站查看
其他类型引用(1)