订单剩余时间读取区域修复

depot_test/仓库识别demo copy.py

@@ -24,17 +24,6 @@ def 提取特征点(模板):
     return hog_features
 
 
-def 训练knn模型(images, labels):
-    knn_classifier = KNeighborsClassifier(weights="distance", n_neighbors=1, n_jobs=1)
-    knn_classifier.fit(images, labels)
-    return knn_classifier
-
-
-def 保存knn模型(classifier, filename):
-    with lzma.open(filename, "wb") as f:
-        pickle.dump(classifier, f)
-
-
 class DepotMatcher:
     def __init__(
         self,
@@ -154,6 +143,73 @@ class DepotMatcher:
 
         return self
 
+    def 训练并保存knn模型(self, images, labels, filename):
+        """训练并保存KNN模型"""
+        knn_classifier = KNeighborsClassifier(
+            weights="distance", n_neighbors=1, n_jobs=1
+        )
+        knn_classifier.fit(images, labels)
+
+        with lzma.open(filename, "wb") as f:
+            pickle.dump(knn_classifier, f)
+
+        return knn_classifier
+
+    def 训练knn模型(self, 模型保存路径="depot_knn_model.xz"):
+        """训练并保存KNN模型"""
+        if not self.refs:
+            print("错误：请先加载参考图像！")
+            return None
+
+        # 准备训练数据
+        images = []
+        labels = []
+
+        for name, img_array in self.refs.items():
+            # 提取HOG特征
+            features = 提取特征点(img_array)
+            images.append(features)
+            labels.append(name)
+
+        # 训练模型
+        self.knn_model = self.训练并保存knn模型(images, labels, 模型保存路径)
+        print(f"KNN模型训练完成，已保存到: {模型保存路径}")
+        return self
+
+    def 使用knn预测(self, 测试图像):
+        """使用训练好的KNN模型进行预测"""
+        if not hasattr(self, "knn_model"):
+            print("错误：请先训练KNN模型！")
+            return None, None
+
+        # 提取测试图像的特征
+        features = 提取特征点(测试图像)
+
+        # 预测
+        预测结果 = self.knn_model.predict([features])
+        置信度 = self.knn_model.predict_proba([features])
+
+        return 预测结果[0], 置信度[0]
+
+    def match_knn(self):
+        """KNN匹配方法"""
+        if not hasattr(self, "knn_model"):
+            print("错误：请先训练KNN模型！")
+            return self
+
+        self.knn_results = []
+
+        for idx, gray_roi_np, _ in self.rois:
+            # 使用KNN进行预测
+            预测名称, 置信度 = self.使用knn预测(gray_roi_np)
+
+            # 获取最高置信度的值
+            max_conf = np.max(置信度) if 置信度 is not None else 0
+
+            self.knn_results.append((idx, 预测名称, max_conf))
+
+        return self
+
     def match_template(self, threshold=0.2):
         """模板匹配方法"""
         self.template_results = []
@@ -257,7 +313,7 @@ class DepotMatcher:
         return self
 
     def display_results(self):
-        """可视化匹配结果，直接使用内存中的图像数据"""
+        """可视化匹配结果，现在包含KNN结果"""
         ROW_LIMIT = 9
 
         # 获取一个参考图像的尺寸作为空白图像的基础
@@ -271,14 +327,13 @@ class DepotMatcher:
         current_row_width = 0
         max_row_height = 0
 
-        # self.rois 现在包含 (idx, gray_roi_data, color_roi_data)
         for idx, gray_roi_data, color_roi_data in self.rois:
             color_roi_img = Image.fromarray(
                 cv2.cvtColor(color_roi_data, cv2.COLOR_BGR2RGB)
             )
-
             gray_roi_img = Image.fromarray(gray_roi_data).convert("L").convert("RGB")
 
+            # 获取模板匹配结果
             t_res_name = next(
                 (name for i, name, val in self.template_results if i == idx), None
             )
@@ -291,6 +346,7 @@ class DepotMatcher:
                 else blank_img_pil.copy()
             )
 
+            # 获取SSIM匹配结果
             s_res_details = next((d for d in self.ssim_results if d[0] == idx), None)
             s_res_name = s_res_details[1] if s_res_details else None
             s_ref_img = (
@@ -299,18 +355,32 @@ class DepotMatcher:
                 else blank_img_pil.copy()
             )
 
+            # 获取KNN匹配结果
+            k_res_details = next(
+                (d for d in getattr(self, "knn_results", []) if d[0] == idx), None
+            )
+            k_res_name = k_res_details[1] if k_res_details else None
+            k_ref_img = (
+                Image.fromarray(self.refs[k_res_name]).convert("RGB")
+                if k_res_name and k_res_name in self.refs
+                else blank_img_pil.copy()
+            )
+            k_res_val = k_res_details[2] if k_res_details else 0
+
             # 计算组合尺寸
             combined_width = (
                 color_roi_img.width
                 + gray_roi_img.width
                 + t_ref_img.width
                 + s_ref_img.width
+                + k_ref_img.width
             )
             combined_height = max(
                 color_roi_img.height,
                 gray_roi_img.height,
                 t_ref_img.height,
                 s_ref_img.height,
+                k_ref_img.height,
             )
 
             # 创建组合图像
@@ -330,24 +400,26 @@ class DepotMatcher:
             x_offset += t_ref_img.width
 
             combined.paste(s_ref_img, (x_offset, 0))
+            x_offset += s_ref_img.width
+
+            combined.paste(k_ref_img, (x_offset, 0))
+            x_offset += k_ref_img.width
 
             # 添加标注
             draw = ImageDraw.Draw(combined)
-
             font = ImageFont.truetype("msyh.ttc", 16)
 
             label = (
-                f"ROI {idx} SSIM/H/E:{s_res_details[3]:.2f}/{s_res_details[4]:.2f}/{s_res_details[5]:.2f}\n"
+                f"ROI {idx}\n"
                 f"T: {t_res_name or 'None'} ({t_res_val:.2f})\n"
-                f"S: {s_res_details[1] or 'None'} ({s_res_details[2]:.2f})\n"
+                f"S: {s_res_name or 'None'} ({s_res_details[2]:.2f})\n"
-            )
+                f"K: {k_res_name or 'None'} ({k_res_val:.2f})"
-            # 文本颜色：如果模板匹配和SSIM匹配结果一致，则为红，否则为品红
-            text_color = (
-                (255, 0, 0) if t_res_name == s_res_details[1] else (255, 0, 255)
             )
 
-            # 调整标签位置到模板匹配参考图的左上角
+            # 文本颜色：如果三种方法结果一致则为红色，否则为黑色
-            # (起始 X 坐标是彩色ROI和灰度ROI的宽度之和)
+            text_color = (
+                (255, 0, 0) if t_res_name == s_res_name == k_res_name else (0, 0, 0)
+            )
 
             draw.text(
                 (color_roi_img.width, gray_roi_img.height),
@@ -372,12 +444,11 @@ class DepotMatcher:
                     row_img.paste(img, (x, 0))
                     x += img.width
                 combined_images.append(row_img)
-                # 重置行变量
                 current_row_images = []
                 current_row_width = 0
                 max_row_height = 0
 
-        # 处理最后一行（如果未满 ROW_LIMIT）
+        # 处理最后一行
         if current_row_images:
             row_img = Image.new(
                 "RGB", (current_row_width, max_row_height), (255, 255, 255)
@@ -388,44 +459,35 @@ class DepotMatcher:
                 x += img.width
             combined_images.append(row_img)
 
-        # 如果没有生成任何图像行，则退出
-        if not combined_images:
-            print("没有生成任何结果图像行。")
-            return self
-
         # 生成最终图像
+        if combined_images:
             total_height = sum(img.height for img in combined_images)
-        # 使用第一行的宽度作为所有行的最大宽度（或计算所有行的最大宽度）
+            max_width = max(img.width for img in combined_images)
-        max_width = combined_images[0].width if combined_images else 0
+            final_img = Image.new("RGB", (max_width, total_height), (255, 255, 255))
-        # 如果行宽不一致，可能需要调整逻辑或使用最大宽度
-        final_img = Image.new(
-            "RGB",
-            (max_width, total_height),
-            (255, 255, 255),
-        )
 
             y = 0
             for img in combined_images:
-            # 确保粘贴时宽度不超过 final_img 的宽度
                 final_img.paste(img, (0, y))
                 y += img.height
 
-        # 保存最终结果图像
             output_path = "depot_test/output/matches_all.png"
-        os.makedirs(os.path.dirname(output_path), exist_ok=True)  # 确保目录存在
+            os.makedirs(os.path.dirname(output_path), exist_ok=True)
             final_img.save(output_path)
             print(f"结果图像已保存至: {output_path}")
+
         return self
 
 
 if __name__ == "__main__":
     # 使用示例
     matcher = DepotMatcher()
+
     (
         matcher.load_references()
         .detect_and_crop()
         .match_template(threshold=0.2)
         .match_ssim(threshold=0.05)
+        .match_knn()
         .display_results()
     )
 

depot_test/仓库识别demo.py

@@ -150,9 +150,13 @@ def match_ssim(rois, refs, thresh):
         dynamic_thresh = thresh * (1 + 0.3 * roi_complexity)
 
         if max_combined_score >= dynamic_thresh:
-            results.append((idx, best_match, max_combined_score, best_ssim, best_hist, best_edge))
+            results.append(
+                (idx, best_match, max_combined_score, best_ssim, best_hist, best_edge)
+            )
         else:
-            results.append((idx, None, max_combined_score, best_ssim, best_hist, best_edge))
+            results.append(
+                (idx, None, max_combined_score, best_ssim, best_hist, best_edge)
+            )
 
     return results
 
@@ -281,11 +285,21 @@ if __name__ == "__main__":
         t_val = next((val for i, name, val in results_template if i == idx), 0)
 
         # SSIM results - now includes detailed metrics
-        s_res = next((name for i, name, val, ssim, hist, edge in results_ssim if i == idx), None)
+        s_res = next(
-        s_val = next((val for i, name, val, ssim, hist, edge in results_ssim if i == idx), 0)
+            (name for i, name, val, ssim, hist, edge in results_ssim if i == idx), None
-        s_ssim = next((ssim for i, name, val, ssim, hist, edge in results_ssim if i == idx), 0)
+        )
-        s_hist = next((hist for i, name, val, ssim, hist, edge in results_ssim if i == idx), 0)
+        s_val = next(
-        s_edge = next((edge for i, name, val, ssim, hist, edge in results_ssim if i == idx), 0)
+            (val for i, name, val, ssim, hist, edge in results_ssim if i == idx), 0
+        )
+        s_ssim = next(
+            (ssim for i, name, val, ssim, hist, edge in results_ssim if i == idx), 0
+        )
+        s_hist = next(
+            (hist for i, name, val, ssim, hist, edge in results_ssim if i == idx), 0
+        )
+        s_edge = next(
+            (edge for i, name, val, ssim, hist, edge in results_ssim if i == idx), 0
+        )
 
         print(
             f"ROI {idx}:\n"

depot_test/仓库识别模版匹配.py

@@ -1,11 +1,13 @@
+import json
 import os
+from datetime import datetime
+from multiprocessing import Pool
+
 import cv2
 import numpy as np
-from datetime import datetime
-from skimage.metrics import structural_similarity
 from PIL import Image
-import json
+from skimage.metrics import structural_similarity
-from multiprocessing import Pool
+
 now = datetime.now()
 # 配置路径
 REF_DIR = r"depot_test\output/test/origin"

depot_test/水平拼接stitch.py

@@ -1,8 +1,10 @@
-import cv2
 import glob
 import os
 from datetime import datetime
 
+import cv2
+
+
 def stitch_panorama(input_dir, output_dir):
     os.makedirs(output_dir, exist_ok=True)
     image_paths = sorted(glob.glob(os.path.join(input_dir, "*.*")))
@@ -26,6 +28,7 @@ def stitch_panorama(input_dir, output_dir):
     else:
         print(f"拼接失败，错误码：{status}")
 
+
 if __name__ == "__main__":
     start = datetime.now()
     stitch_panorama(r"depot_test\Test", r"depot_test\output")

mower/solvers/infra/get_order_remaining_time.py

@@ -56,7 +56,7 @@ class GetOrderRemainingTimeSolver(BaseSolver, BaseMixin):
             elif pos := self.find("bill_accelerate"):
                 scope = (70, -203), (194, -170)
                 scope = sa(scope, pos[0])
-                if res := self.read_remain_time(pos):
+                if res := self.read_remain_time(scope):
                     self.res = res
                     return True