零、NEED
YOLOv8中文官网 https://docs.ultralytics.com/zh
YOLOv8Github https://github.com/ultralytics/ultralytics
YOLOv8n模型 https://github.com/ultralytics/assets/releases/download/v8.1.0/yolov8n.pt
若打不开Github,可使用fastgithub
一、安装ultralytics库
Terminal(Cli、CMD)运行以下代码,使用清华源加速安装ultralytics
pip install ultralytics -i https://pypi.tuna.tsinghua.edu.cn/simple
二、运行代码
需要把YOLOv8模型放置在py文件同一目录下
若不清楚py文件目录,可右键复制路径
方式一:检测结果生成在run目录下
from ultralytics import YOLO
from PIL import Image
import cv2
model = YOLO("yolov8n.pt")
# from PIL
im1 = "https://5b0988e595225.cdn.sohucs.com/images/20180429/7f9b59894ceb4f409578277d821a6809.png" ##这里放图片网址,目录格式是 im1 = Image.open(" ")
results = model.predict(source=im1, save=True) # save plotted images
检测结果在run目录下
方式二:检测结果直接弹窗
from ultralytics import YOLO
import cv2
# 加载预训练模型
model = YOLO("yolov8n.pt", task='detect')
# model = YOLO("yolov8n.pt") task参数也可以不填写,它会根据模型去识别相应任务类别
# 检测图片
results = model("https://5b0988e595225.cdn.sohucs.com/images/20180429/7f9b59894ceb4f409578277d821a6809.png") #这里放图片网址,也可以是目录格式
res = results[0].plot()
cv2.imshow("YOLOv8 Inference", res)
cv2.waitKey(0)
结果: