首页|基于计算机软件技术在大数据时代的应用研究

基于计算机软件技术在大数据时代的应用研究

扫码查看
在当前大数据时代背景下,计算机软件技术的迅速发展已经引领人类进入了一个信息爆炸的新时代.在交通目标识别问题中,也广泛应用了计算机软件技术,然而低光照交通环境会显著降低图像的质量,进而影响识别的准确率.为了解决这一问题,研究提出Yolov4的改进算法CLAHE-GhostNet-CBAM-Yolov4(CGC-Yolov4).研究在输入模块中对输入图像采用CLAHE图像增强算法,用GhostNet替代Yolov4主干网络,在瓶颈网络的末端添加了注意力机制.性能测试的结果表明在行人目标识别中,研究提出的模型Fl得分在第12次迭代达到91%,最高可达98%.实验结果表明,相较于传统模型CGC-Yolov4不仅提高了识别效率,也保持了极佳的实时处理能力,证明了其在夜间交通目标识别应用中的实用性和有效性.
Research Based on the Application of Computer Software Technology in the Era of Big Data
In the context of the current big data era,the rapid development of computer software technology has led mankind into a new era of information explosion.In the traffic target recognition problem,computer software technology is also widely used.However,the low-light traffic environ-ment will significantly reduce the quality of the image,which in turn affects the accuracy of recognition.In order to solve this problem,the study proposes an improved algorithm CLAHE-GhostNet-CBAM-Yolov4(CGC-Yolov4)for Yolov4.The study uses the CLAHE image enhancement algorithm for the input image in the input module,replaces the Yolov4 backbone network with GhostNet,and adds the attention mechanism at the end of the bottleneck network.The results of the performance tests show that in pedestrian target recognition,the F1 score of the proposed model reaches 91%up to 98%in the 12th iteration.The experimental results show that compared to the traditional model CGC-Yolov4 not only improves the recognition efficiency,but also maintains excellent real-time processing capability,which proves its practicality and effectiveness in night-time traffic target recognition appli-cations.

intelligent transportation systemdeep learningattention mechanismYOLOv4mobile visual recognition

贾豁然

展开 >

沈阳音乐学院,辽宁沈阳 110000

智能交通系统 深度学习 注意力机制 YOLOv4 移动视觉识别

2024

佳木斯大学学报(自然科学版)
佳木斯大学

佳木斯大学学报(自然科学版)

影响因子:0.159
ISSN:1008-1402
年,卷(期):2024.42(8)