基于改进YOLOv8n的机器人视觉系统目标识别算法研究

Research on Object Recognition Algorithm for Robot Vision System Based on Improved YOLOv8n

针对机器人视觉系统在开阔环境中检测精度低的问题,该文提出了基于改进YOLOv8n的目标检测算法。实验结果显示,在引入多尺度检测头和压缩-激励注意力模块后,YOLOv8n的准确率上升到82.3%,而在引入Ghost卷积后,算法的浮点运算数降低至11.6。同时基于改进YOLOv8n的目标检测算法在水面环境中的检测准确率可达83.6%,高于其他算法。上述结果表明,基于改进YOLOv8n的目标检测算法不仅检测精度高,且计算复杂度低。

In response to the problem of low detection accuracy of robot vision systems in open environments,a target detection algorithm based on improved YOLOv8n is proposed.The experimental results show that with the introduction of a multi-scale detection head and a compression excitation attention module,the accuracy of YOLOv8n increases to 82.3%,while with the introduction of Ghost convolution,the floating-point operations of the algorithm decrease to 11.6.At the same time,the detection accuracy of the object detection algorithm based on improved YOLOv8n in water surface environment can reach 83.6%,which is higher than other algorithms.The above results indicate that the object detection algorithm based on improved YOLOv8n not only has high detection accuracy,but also has low computational complexity.