modify

.gitignore

@@ -12,6 +12,8 @@ result1/
 result/
 *.pyc
 plate/
+wan/
+label_imgs/
 # 不忽略下面指定的文件类型
 !*.cpp
 !*.h

README.md

@@ -23,7 +23,7 @@ python detect_rec_plate.py --detect_model weights/plate_detect.pt  --rec_model w
 
 ## **车牌检测训练**
 
-1. 下载数据集：  [datasets](https://pan.baidu.com/s/1rQ3ScJ_AdroDcKvUNW8Fhw) 提取码：g5wk    数据从CCPD和CRPD数据集中选取并转换的
+1. 下载数据集：  [datasets](https://pan.baidu.com/s/1xa6zvOGjU02j8_lqHGVf0A) 提取码：pi6c 数据从CCPD和CRPD数据集中选取并转换的
    数据集格式为yolo格式：
 
    ```

detect_rec_plate.py

@@ -12,6 +12,9 @@ from plate_recognition.plate_rec import get_plate_result,allFilePath,init_model,
 from plate_recognition.double_plate_split_merge import get_split_merge
 from utils.datasets import letterbox
 from utils.cv_puttext import cv2ImgAddText
+def cv_imread(path):
+    img=cv2.imdecode(np.fromfile(path,dtype=np.uint8),-1)
+    return img
 
 clors = [(255,0,0),(0,255,0),(0,0,255),(255,255,0),(0,255,255)]
 
@@ -72,6 +75,7 @@ def get_plate_rec_landmark(img, xyxy, conf, landmarks, class_num,device,plate_re
     result_dict['plate_no']=plate_number
     result_dict['roi_height']=roi_img.shape[0]
     result_dict['score']=conf
+    result_dict['label']=class_label
     return result_dict
 
 def detect_Recognition_plate(model, orgimg, device,plate_rec_model,img_size):
@@ -134,9 +138,9 @@ def draw_result(orgimg,dict_list):
 
 if __name__ == '__main__':
     parser = argparse.ArgumentParser()
-    parser.add_argument('--detect_model', nargs='+', type=str, default='weights/plate_detect.pt', help='model.pt path(s)')
+    parser.add_argument('--detect_model', nargs='+', type=str, default='runs/train/yolov75/weights/best.pt', help='model.pt path(s)')
     parser.add_argument('--rec_model', type=str, default='weights/plate_rec.pth', help='model.pt path(s)')
-    parser.add_argument('--source', type=str, default='imgs', help='source')  # file/folder, 0 for webcam
+    parser.add_argument('--source', type=str, default='wan', help='source')  # file/folder, 0 for webcam
     # parser.add_argument('--img-size', nargs= '+', type=int, default=640, help='inference size (pixels)')
     parser.add_argument('--img_size', type=int, default=640, help='inference size (pixels)')
     parser.add_argument('--output', type=str, default='result', help='source') 
@@ -156,7 +160,9 @@ if __name__ == '__main__':
     time_b = time.time()
     for pic_ in file_list:
         print(pic_,end=" ")
-        img = cv2.imread(pic_)
+        img = cv_imread(pic_)
+        if img.shape[-1]==4:
+            img=cv2.cvtColor(img,cv2.COLOR_BGRA2BGR)
         # img = my_letter_box(img)
         dict_list=detect_Recognition_plate(model, img, device,plate_rec_model,opt.img_size)
         ori_img=draw_result(img,dict_list)

get_small_pic.py

@@ -0,0 +1,94 @@
+from detect_rec_plate import detect_Recognition_plate,attempt_load,init_model,allFilePath,cv_imread,draw_result,four_point_transform,order_points
+import os
+import argparse
+import torch
+import cv2
+import time
+import shutil
+import numpy as np
+import json
+import re
+pattern_str = "([京津沪渝冀豫云辽黑湘皖鲁新苏浙赣鄂桂甘晋蒙陕吉闽贵粤青藏川宁琼]" \
+              "{1}(([A-HJ-Z]{1}[A-HJ-NP-Z0-9]{5})|([A-HJ-Z]{1}(([DF]{1}[A-HJ-NP-Z0-9]{1}[0-9]{4})|([0-9]{5}[DF]" \
+              "{1})))|([A-HJ-Z]{1}[A-D0-9]{1}[0-9]{3}警)))|([0-9]{6}使)|((([沪粤川云桂鄂陕蒙藏黑辽渝]{1}A)|鲁B|闽D|蒙E|蒙H)" \
+              "[0-9]{4}领)|(WJ[京津沪渝冀豫云辽黑湘皖鲁新苏浙赣鄂桂甘晋蒙陕吉闽贵粤青藏川宁琼·•]{1}[0-9]{4}[TDSHBXJ0-9]{1})" \
+              "|([VKHBSLJNGCE]{1}[A-DJ-PR-TVY]{1}[0-9]{5})"
+
+
+def is_car_number(pattern, string):
+    if re.findall(pattern, string):
+        return True
+    else:
+        return False
+
+if __name__ == '__main__':
+    parser = argparse.ArgumentParser()
+    parser.add_argument('--detect_model', nargs='+', type=str, default=r'runs/train/yolov75/weights/best.pt', help='model.pt path(s)')
+    parser.add_argument('--rec_model', type=str, default=r'weights/plate_rec.pth', help='model.pt path(s)')
+    parser.add_argument('--source', type=str, default=r'/mnt/Gpan/BaiduNetdiskDownload/VehicleColour/VehicleColour/', help='source')  # file/folder, 0 for webcam
+    # parser.add_argument('--source', type=str, default=r'test2', help='source')
+    # parser.add_argument('--img-size', nargs= '+', type=int, default=640, help='inference size (pixels)')
+    parser.add_argument('--img_size', type=int, default=640, help='inference size (pixels)')
+    parser.add_argument('--output', type=str, default=r'/mnt/Gpan/BaiduNetdiskDownload/VehicleColour/result1', help='source') 
+    parser.add_argument('--kpt-label', type=int, default=4, help='number of keypoints')
+    device  =torch.device("cuda" if torch.cuda.is_available() else "cpu")
+    # device = torch.device("cpu")
+    opt = parser.parse_args()
+    print(opt)
+    model = attempt_load(opt.detect_model, map_location=device)
+    # torch.save()
+    plate_rec_model=init_model(device,opt.rec_model) 
+    if not os.path.exists(opt.output):
+        os.mkdir(opt.output)
+    file_list=[]
+    index_1=0
+    index_small=0
+    # error_path =r"E:\study\plate\data\@shaixuan\train\pic"
+    allFilePath(opt.source,file_list)
+    time_b = time.time()
+   
+    for pic_ in file_list:
+        try:
+            image_name = os.path.basename(pic_)
+            # ori_plate = image_name.split("_")[0]
+            # ori_plate = image_name.split(".")[0]
+            flag = 1
+            index_1+=1
+            label_dict_str=""
+            lable_dict={}
+            # label_dict={}                                 
+            print(index_small,index_1,pic_)
+            img = cv_imread(pic_)
+            if img is None:
+                continue
+            if img.shape[-1]==4:
+                img=cv2.cvtColor(img,cv2.COLOR_BGRA2BGR)
+            # img = my_letter_box(img)
+            count=0
+            dict_list=detect_Recognition_plate(model, img, device,plate_rec_model,opt.img_size)
+            for result_ in dict_list:
+                if not result_:
+                    continue
+                index_small+=1
+                landmarks=result_['landmarks']
+                landmarks_np = np.array(landmarks).reshape(-1,2)
+                img_roi = four_point_transform(img,landmarks_np)
+                plate_no= result_['plate_no']
+                if len(plate_no)<6 or not is_car_number(pattern_str,plate_no):
+                    continue
+                height = result_['roi_height']
+                # if height<48:
+                #     continue
+                pic_name =plate_no+"_"+str(index_small)+".jpg"
+                label = result_["label"]
+                
+
+                roi_save_folder = os.path.join(opt.output,str(label))
+                if not os.path.exists(roi_save_folder):
+                    os.mkdir(roi_save_folder)
+                roi_path = os.path.join(roi_save_folder,pic_name)
+                cv2.imencode('.jpg',img_roi)[1].tofile(roi_path)
+        except:
+            print("error!")
+  
+