基于LGC的反残差目标检测算法

Inverted residual target detection algorithm based on LGC

基于深度学习的目标检测是计算机视觉领域的研究热点,目前主流的目标检测模型大多通过增加网络深度和宽度以获得更好的检测效果,但容易导致参数量增加、检测速度降低的问题。为兼顾检测精度与速度,借鉴Ghost卷积和分组卷积的轻量化思想,提出了一种高效的轻量级Ghost卷积(LGC)模型,以采用更少的参数获得更多的特征图。在该卷积模型的基础上引入反残差结构重新设计了CSPDarkNet53,生成了一种基于LGC的反残差特征提取网络,以提高网络对全局特征信息的提取能力。使用反残差特征提取网络替换YOLOv4的骨干网络,辅以深度可分离卷积进一步减少参数,提出了一种反残差目标检测算法,以提升目标检测的整体性能。实验结果表明:相比于主流的目标检测算法,所提算法在检测精度相当的前提下,模型参数量和检测速度具有明显的优势。

Target detection based on deep learning is a research hotspot in computer vision.Although existing mainstream detection models usually increase the depth and width of the network to acquire better detection results,it is unamiable to suffer from parameters increasing and detection rate decreasing.To address this problem,an efficient lightweight Ghost convolution(LGC)model,which aims to balance the detection accuracy and speed,and obtain more feature maps with fewer parameters,was proposed by referring to the lightweight idea of Ghost convolution and group convolution.CSPDarkNet53 that was redesigned with the above convolution and an inverted residual structure was introduced to generate an inverted residual feature extraction network to improve the global feature information extraction capability of the model.On this basis,the inverted residual feature extraction network was used as the backbone network of YOLOv4,and depthwise separable convolution was used to reduce the parameters.To improve the overall performance of the algorithm,an inverted residual target detection algorithm was proposed.Experimental results show that compared with the current mainstream target detection algorithm,the proposed algorithm has prominent advantages in the number of model parameters and detection speed under the premise of similar detection accuracy.