dl_anomaly_detection_workflow

dl_anomaly_detection_workflow(Halcon独有的，只需要训练OK，就可以做检测)*****
*
* Deep learning anomaly detection example.
*
* This example demonstrates(表明，显示) the general workflow for anomaly detection based on deep learning.
*
* The workflow typically(典型的，经典的，常常的) consists(由.......组成) of four steps which are described here:

* 1. Dataset(数据集) preprocessing.
* 2. Training of the model and threshold estimation on 'ok' images.（训练模型，分类）
* 3. Qualitative evaluation of the trained model.对训练模型的定性评估。
* 4. Inference on new images.（推理）

××××××××××××××××××××××××
dev_update_off ()
（GSZ：Switch dev_update_pc, dev_update_var and dev_update_window to 'off'.）
（dev_update_pc----        Switches the update of the PC during program execution on or off.）
（dev_update_var---        Switches the update of the variable window during program execution on or off.）
（dev_update_window -- Switches the automatic output of iconic output objects into the graphics window during program execution on or off.）
***
set_system ('seed_rand', 73)  
（GSZ：设定随机种子）

**
* Set path to data.
get_image_dir (HalconImages)


（GSZ：HalconImages 是一个系统变量，代表MVT提供的图像的素材的路径）