基于DCP-Imp CycleGAN CenterNet去雾算法的交通标志检测

Traffic Sign Detection Based on DCP-Imp CycleGAN CenterNet Dehazing Algorithm

针对雾天环境下对小型交通标志检测效果不佳的问题,提出了一种基于DCP-Imp CycleGAN与CenterNet融合去雾的交通标志检测方法。该方法在预处理模块将DCP算法嵌入到优化后的CycleGAN网络框架中,对图像高质量的纹理信息细化处理,再通过感知融合模块,获得更易被识别且自然的无雾图像;为了进一步提高对小型交通标志的识别能力,对CenterNet中的残差块进行了轻量化设计,同时引入了CBAM注意力机制和FPN特征融合模块,进而减少了有效特征信息的丢失。实验结果表明,改进算法能有效解决图像去雾中色差明显和不清晰的问题,在CCTSDB数据集上实验评估得到的mAP较CenterNet提升了5.48%,FPS提升了4帧,有效解决了雾天环境下对小型交通标志的漏检、误检问题。

To address the difficulty of traffic sign detection in haze environment,an improved fusion dehazing method based on DCP-Imp CycleGAN and CenterNet is proposed,in which the DCP algorithm is embedded in the optimized CycleGAN network framework to refine the high-quality texture information.Then a perception fusion module is applied to obtain a recognizable and natural dehazed image.In order to further improve the recognition ability of small traffic signs,a lightweight design is adopted for the residual blocks in CenterNet,and the CBAM attention mechanism and FPN feature fusion module are introduced to reduce the loss of effective feature information.The experimental results show that the algorithm can effectively solve the problem of unclearness and color difference in image dehazing.The mAP evaluated in CCTSDB data set is increased by 5.48%,and the FPS 4 frames.It achieves good results of small traffic sign detection in foggy environment.