59 lines
1.3 KiB
Markdown
59 lines
1.3 KiB
Markdown
|
# yolov11_cls_inference README
|
|||
|
|
|
|||
|
|
## 概述
|
|||
|
|
该模块用于对米厂输入图像执行二分类推理，用于判断机械臂夹爪是否夹紧。
|
|||
|
|
|
|||
|
|
类别定义：
|
|||
|
|
|
|||
|
|
0 → 夹具夹紧 (False)
|
|||
|
|
1 → 夹具打开 (True)
|
|||
|
|
|
|||
|
|
rknn模型只加载一次，复用全局实例，提高推理效率。
|
|||
|
|
|
|||
|
|
## 调用示例
|
|||
|
|
|
|||
|
|
您可以直接调用 yolov11_cls_inference 函数，以便集成到其他项目中：
|
|||
|
|
|
|||
|
|
示例 1: 单张图片推理
|
|||
|
|
|
|||
|
|
```bash
|
|||
|
|
from main_cls import yolov11_cls_inference
|
|||
|
|
import cv2
|
|||
|
|
|
|||
|
|
# 读取图像
|
|||
|
|
bgr_image = cv2.imread("11.jpg")
|
|||
|
|
rgb_image = cv2.cvtColor(bgr_image, cv2.COLOR_BGR2RGB)
|
|||
|
|
|
|||
|
|
# 调用推理函数
|
|||
|
|
class_id, bool_value = yolov11_cls_inference(
|
|||
|
|
model_path="yolov11_cls.rknn",
|
|||
|
|
raw_image=rgb_image,
|
|||
|
|
target_size=(640, 640)
|
|||
|
|
)
|
|||
|
|
|
|||
|
|
print(f"类别ID: {class_id}, 布尔值: {bool_value}")
|
|||
|
|
|
|||
|
|
```
|
|||
|
|
|
|||
|
|
示例 2: 多次推理（复用模型）
|
|||
|
|
```bash
|
|||
|
|
|
|||
|
|
from main_cls import init_rknn_model, yolov11_cls_inference_once
|
|||
|
|
import cv2
|
|||
|
|
|
|||
|
|
# 初始化一次模型
|
|||
|
|
rknn_model = init_rknn_model("cls_rk3568.rknn")
|
|||
|
|
|
|||
|
|
# 读取图像
|
|||
|
|
bgr_image = cv2.imread("12.jpg")
|
|||
|
|
rgb_image = cv2.cvtColor(bgr_image, cv2.COLOR_BGR2RGB)
|
|||
|
|
|
|||
|
|
# 使用已加载模型进行推理
|
|||
|
|
class_id, bool_value = yolov11_cls_inference_once(rknn_model, rgb_image)
|
|||
|
|
|
|||
|
|
if bool_value:
|
|||
|
|
print("夹具夹紧")
|
|||
|
|
else:
|
|||
|
|
print("夹具打开")
|
|||
|
|
```
|