'commit'

Apps/TeLaiDian/Config/Setting.py

@@ -18,3 +18,4 @@ WAIT_AFTER_SCROLL = 2.5
 SAFE_EXCLUDE_RATIO = 0.55  # 大幅增加排除比例，确保从过滤器下方开始识别
 BOTTOM_SAFE_EXCLUDE_RATIO = 0.12 
 MIN_CARD_HEIGHT = 250 # 增加最小高度要求，确保卡片信息完整（特来电卡片较长，约300px）
+DETAIL_SCROLL_DISTANCE_RATIO = 0.9 # 详情页滑动距离比例，确保露出价格按钮

Apps/TeLaiDian/Crawler.py

@@ -10,7 +10,7 @@ from Apps.TeLaiDian.Service import TeLaiDianService
 from Apps.TeLaiDian.Config.Setting import (
     SCROLL_DISTANCE_RATIO, WAIT_AFTER_SCROLL, MAX_STATIONS_COUNT,
     SAFE_EXCLUDE_RATIO, BOTTOM_SAFE_EXCLUDE_RATIO, WAIT_DETAIL_PAGE_LOAD,
-    WAIT_BACK_TO_LIST
+    WAIT_BACK_TO_LIST, DETAIL_SCROLL_DISTANCE_RATIO
 )
 from Core.BaseCrawler import BaseCrawler
 import uiautomator2 as u2
@@ -115,9 +115,9 @@ class TeLaiDianCrawler(BaseCrawler):
         logger.info(f"详情页基础信息识别完成: {station_name} | {address}")
 
         # 2. 向上滑动以露出价格按钮
-        logger.info("执行滑动操作以显示价格按钮...")
+        logger.info(f"执行滑动操作以显示价格按钮 (距离比例: {DETAIL_SCROLL_DISTANCE_RATIO})...")
-        # 从屏幕中间向上滑动，scale=0.6 表示滑动距离约为屏幕高度的 60%
+        # 从屏幕中间向上滑动
-        d.swipe_ext("up", scale=0.6)
+        d.swipe_ext("up", scale=DETAIL_SCROLL_DISTANCE_RATIO)
         await asyncio.sleep(1.5)
 
         # 3. 点击“价格信息”按钮 (jgxx.jpg)

Apps/TeLaiDian/TestDetailCV.py

@@ -0,0 +1,97 @@
+# coding=utf-8
+import os
+import sys
+import cv2
+import numpy as np
+
+# 将项目根目录添加到 sys.path
+project_root = os.path.dirname(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+if project_root not in sys.path:
+    sys.path.append(project_root)
+
+from Apps.TeLaiDian import Kit
+
+def detect_price_areas_cv(image_path):
+    """
+    专门用于详情页识别价格点击区域的测试函数
+    """
+    img = Kit.read_image(image_path)
+    if img is None:
+        print("错误: 无法读取图片")
+        return []
+    
+    h, w = img.shape[:2]
+    gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
+    
+    # 1. 转换为 HSV 空间，方便寻找橘红色价格
+    hsv = cv2.cvtColor(img, cv2.COLOR_BGR2HSV)
+    
+    # 橘红色的 HSV 范围 (大致)
+    lower_orange = np.array([0, 150, 150])
+    upper_orange = np.array([20, 255, 255])
+    mask = cv2.inRange(hsv, lower_orange, upper_orange)
+    
+    # 2. 对掩码进行膨胀，连接数字
+    kernel = cv2.getStructuringElement(cv2.MORPH_RECT, (50, 20))
+    dilated = cv2.dilate(mask, kernel)
+    
+    # 3. 寻找轮廓
+    contours, _ = cv2.findContours(dilated, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
+    
+    detected_areas = []
+    
+    for cnt in contours:
+        x, y, cw, ch = cv2.boundingRect(cnt)
+        
+        # 橘红色价格通常在屏幕上半部，且宽度适中
+        if 150 < y < h * 0.6 and cw > 100:
+            # 我们找到了价格数字。用户说要点击的是这个区域。
+            # 我们可以适当扩大这个区域，或者直接取其中心。
+            padding = 20
+            detected_areas.append([
+                max(0, x - padding), 
+                max(0, y - padding), 
+                min(w, x + cw + padding), 
+                min(h, y + ch + padding)
+            ])
+
+    # 按 X 轴排序，找到最左边的价格
+    detected_areas.sort(key=lambda b: b[0])
+    return detected_areas
+
+def run_test(image_path):
+    print(f"分析图片: {image_path}")
+    
+    # 获取所有可能的区域
+    bboxes = detect_price_areas_cv(image_path)
+    print(f"检测到 {len(bboxes)} 个潜在区域:")
+    for i, box in enumerate(bboxes):
+        print(f"  区域 {i}: {box} (w={box[2]-box[0]}, h={box[3]-box[1]})")
+    
+    # 生成 _vl.jpg (仅框)
+    vl_path = image_path.replace(".jpg", "_vl.jpg")
+    img_vl = Kit.read_image(image_path)
+    for i, box in enumerate(bboxes):
+        cv2.rectangle(img_vl, (box[0], box[1]), (box[2], box[3]), (0, 255, 0), 2)
+        cv2.putText(img_vl, str(i), (box[0], box[1]-5), cv2.FONT_HERSHEY_SIMPLEX, 0.8, (0, 255, 0), 2)
+    Kit.save_image(vl_path, img_vl)
+    print(f"已生成标注图: {vl_path}")
+
+    # 生成 _flag.jpg (框 + 中心点)
+    flag_path = image_path.replace(".jpg", "_flag.jpg")
+    img_flag = Kit.read_image(image_path)
+    for i, box in enumerate(bboxes):
+        cv2.rectangle(img_flag, (box[0], box[1]), (box[2], box[3]), (0, 255, 0), 2)
+        center_x = (box[0] + box[2]) // 2
+        center_y = (box[1] + box[3]) // 2
+        cv2.circle(img_flag, (center_x, center_y), 10, (0, 0, 255), -1)
+        cv2.putText(img_flag, f"P{i}", (box[0], box[1]-5), cv2.FONT_HERSHEY_SIMPLEX, 0.8, (0, 0, 255), 2)
+    Kit.save_image(flag_path, img_flag)
+    print(f"已生成人工核对图: {flag_path}")
+
+if __name__ == "__main__":
+    test_image = r"d:\dsWork\aiData\Output\Screenshot_20260114_075758.jpg"
+    if os.path.exists(test_image):
+        run_test(test_image)
+    else:
+        print(f"错误: 找不到测试图片 {test_image}")