diff --git a/.DS_Store b/.DS_Store index d4d83e6..aa13957 100644 Binary files a/.DS_Store and b/.DS_Store differ diff --git a/logs/robot_2025-05-19.log b/logs/robot_2025-05-19.log index 41c9546..ca67f07 100644 --- a/logs/robot_2025-05-19.log +++ b/logs/robot_2025-05-19.log @@ -55,3 +55,13 @@ 2025-05-19 20:50:40 | DEBUG | utils.log_helper - 🐞 显示边缘斜率和中线交点 2025-05-19 20:50:41 | INFO | utils.log_helper - ℹ️ 保存横向边缘检测结果图像到: logs/image/horizontal_edge_20250519_205041_583040.jpg 2025-05-19 20:50:41 | INFO | utils.log_helper - ℹ️ 横向边缘检测结果: {'timestamp': '20250519_205041_583040', 'edge_point': (92, 1077), 'distance_to_center': -868, 'slope': -0.07008086253369272, 'distance_to_bottom': 63.83018867924534, 'intersection_point': (960, 1016), 'is_truncated': False} +2025-05-19 21:24:41 | DEBUG | utils.log_helper - 🐞 步骤1: 原始图像已加载 +2025-05-19 21:24:42 | DEBUG | utils.log_helper - 🐞 步骤2: 创建黄色掩码 +2025-05-19 21:24:43 | DEBUG | utils.log_helper - 🐞 步骤3: 提取黄色部分 +2025-05-19 21:24:44 | DEBUG | utils.log_helper - 🐞 检测底部和顶部边缘点 +2025-05-19 21:24:45 | DEBUG | utils.log_helper - 🐞 步骤4: 边缘检测 +2025-05-19 21:24:46 | DEBUG | utils.log_helper - 🐞 步骤5: 检测到 18 条直线 +2025-05-19 21:24:47 | DEBUG | utils.log_helper - 🐞 步骤6: 找到 3 条水平线 +2025-05-19 21:24:48 | DEBUG | utils.log_helper - 🐞 显示边缘斜率和中线交点 +2025-05-19 21:24:49 | INFO | utils.log_helper - ℹ️ 保存横向边缘检测结果图像到: logs/image/horizontal_edge_20250519_212449_369767.jpg +2025-05-19 21:24:49 | INFO | utils.log_helper - ℹ️ 横向边缘检测结果: {'timestamp': '20250519_212449_369767', 'edge_point': (540, 907), 'distance_to_center': -420, 'slope': -0.12202852614896989, 'distance_to_bottom': 224.25198098256737, 'intersection_point': (960, 855), 'score': 0.5611365011519962, 'valid': True, 'reason': ''} diff --git a/logs/robot_2025-05-20.log b/logs/robot_2025-05-20.log new file mode 100644 index 0000000..a5017c5 --- /dev/null +++ b/logs/robot_2025-05-20.log @@ -0,0 +1,10 @@ +2025-05-20 09:24:31 | DEBUG | utils.log_helper - 🐞 步骤1: 原始图像已加载 +2025-05-20 09:24:32 | DEBUG | utils.log_helper - 🐞 步骤2: 创建黄色掩码 +2025-05-20 09:24:33 | DEBUG | utils.log_helper - 🐞 步骤3: 提取黄色部分 +2025-05-20 09:24:34 | DEBUG | utils.log_helper - 🐞 检测底部和顶部边缘点 +2025-05-20 09:24:35 | DEBUG | utils.log_helper - 🐞 步骤4: 边缘检测 +2025-05-20 09:24:36 | DEBUG | utils.log_helper - 🐞 步骤5: 检测到 18 条直线 +2025-05-20 09:24:37 | DEBUG | utils.log_helper - 🐞 步骤6: 找到 3 条水平线 +2025-05-20 09:24:38 | DEBUG | utils.log_helper - 🐞 显示边缘斜率和中线交点 +2025-05-20 09:24:39 | INFO | utils.log_helper - ℹ️ 保存横向边缘检测结果图像到: logs/image/horizontal_edge_20250520_092439_823936.jpg +2025-05-20 09:24:39 | INFO | utils.log_helper - ℹ️ 横向边缘检测结果: {'timestamp': '20250520_092439_823936', 'edge_point': (540, 907), 'distance_to_center': -420, 'slope': -0.12202852614896989, 'distance_to_bottom': 224.25198098256737, 'intersection_point': (960, 855), 'score': 0.5611365011519962, 'valid': True, 'reason': ''} diff --git a/res/path/test-2.jpg b/res/path/test-2.jpg new file mode 100644 index 0000000..3d676ec Binary files /dev/null and b/res/path/test-2.jpg differ diff --git a/test/task-path-track/yellow_track_demo.py b/test/task-path-track/yellow_track_demo.py index 278cfc0..dcc8b2e 100644 --- a/test/task-path-track/yellow_track_demo.py +++ b/test/task-path-track/yellow_track_demo.py @@ -44,7 +44,7 @@ def process_image(image_path, save_dir=None, show_steps=False): def main(): parser = argparse.ArgumentParser(description='黄色赛道检测演示程序') - parser.add_argument('--input', type=str, default='res/path/test-1.jpg', help='输入图像或视频的路径') + parser.add_argument('--input', type=str, default='res/path/test-2.jpg', help='输入图像或视频的路径') parser.add_argument('--output', type=str, default='res/path/test/result_image_20250514_024313.png', help='输出结果的保存路径') parser.add_argument('--type', type=str, choices=['image', 'video'], help='输入类型，不指定会自动检测') parser.add_argument('--show', default=True, action='store_true', help='显示处理步骤') diff --git a/utils/detect_track.py b/utils/detect_track.py index 2a0386d..8ad3bf3 100644 --- a/utils/detect_track.py +++ b/utils/detect_track.py @@ -42,6 +42,11 @@ def detect_horizontal_track_edge(image, observe=False, delay=1000, save_log=True bottom_bound = height top_bound = height - search_height + # 定义合理的值范围 + valid_y_range = (height * 0.5, height) # 有效的y坐标范围（下半部分图像） + max_slope = 0.15 # 最大允许斜率（接近水平） + min_line_length = width * 0.2 # 最小线长度 + if observe: debug("步骤1: 原始图像已加载", "加载") search_region_img = img.copy() @@ -118,8 +123,8 @@ def detect_horizontal_track_edge(image, observe=False, delay=1000, save_log=True cv2.imshow("边缘检测", edges) cv2.waitKey(delay) - # 使用霍夫变换检测直线 - lines = cv2.HoughLinesP(edges, 1, np.pi/180, threshold=30, + # 使用霍夫变换检测直线，降低阈值以检测更多线段 + lines = cv2.HoughLinesP(edges, 1, np.pi/180, threshold=25, minLineLength=width*0.1, maxLineGap=30) if lines is None or len(lines) == 0: @@ -148,181 +153,62 @@ def detect_horizontal_track_edge(image, observe=False, delay=1000, save_log=True slope = (y2 - y1) / (x2 - x1) # 筛选接近水平的线 (斜率接近0)，但容许更大的倾斜度 - if abs(slope) < 0.3: + if abs(slope) < max_slope: # 确保线在搜索区域内 if ((left_bound <= x1 <= right_bound and top_bound <= y1 <= bottom_bound) or (left_bound <= x2 <= right_bound and top_bound <= y2 <= bottom_bound)): # 计算线的中点y坐标 mid_y = (y1 + y2) / 2 line_length = np.sqrt((x2-x1)**2 + (y2-y1)**2) - # 保存线段、其y坐标和长度 - horizontal_lines.append((line[0], mid_y, slope, line_length)) + + # 过滤掉短线段和太靠近图像上部的线 + if line_length >= min_line_length and mid_y >= valid_y_range[0]: + # 计算线段在图像中的位置得分（越靠近底部得分越高） + position_score = min(1.0, (mid_y - valid_y_range[0]) / (valid_y_range[1] - valid_y_range[0])) + + # 计算长度得分（越长越好） + length_score = min(1.0, line_length / (width * 0.5)) + + # 计算斜率得分（越水平越好） + slope_score = max(0.0, 1.0 - abs(slope) / max_slope) + + # 计算线段位于图像中央的程度 + mid_x = (x1 + x2) / 2 + center_score = max(0.0, 1.0 - abs(mid_x - center_x) / (width * 0.3)) + + # 计算综合得分 + quality_score = position_score * 0.4 + length_score * 0.3 + slope_score * 0.2 + center_score * 0.1 + + # 保存线段、其y坐标、斜率、长度和质量得分 + horizontal_lines.append((line[0], mid_y, slope, line_length, quality_score)) if not horizontal_lines: if observe: - error("未检测到水平线", "失败") + error("未检测到合格的水平线", "失败") return None, None if observe: debug(f"步骤6: 找到 {len(horizontal_lines)} 条水平线", "处理") h_lines_img = img.copy() for line_info in horizontal_lines: - line, _, slope, _ = line_info + line, _, slope, _, score = line_info x1, y1, x2, y2 = line - cv2.line(h_lines_img, (x1, y1), (x2, y2), (0, 255, 255), 2) - # 显示斜率 - cv2.putText(h_lines_img, f"{slope:.2f}", ((x1+x2)//2, (y1+y2)//2), - cv2.FONT_HERSHEY_SIMPLEX, 0.5, (255, 0, 0), 1) + # 根据得分调整线的颜色，得分越高越绿 + color = (int(255 * (1-score)), int(255 * score), 0) + cv2.line(h_lines_img, (x1, y1), (x2, y2), color, 2) + # 显示斜率和得分 + cv2.putText(h_lines_img, f"{slope:.2f}|{score:.2f}", ((x1+x2)//2, (y1+y2)//2), + cv2.FONT_HERSHEY_SIMPLEX, 0.5, color, 1) cv2.imshow("水平线", h_lines_img) cv2.waitKey(delay) - # 将水平线分为上边缘线和下边缘线（按y坐标排序） - bottom_line = None - top_line = None - is_truncated = False # 标记下方线是否被截断 + # 根据质量得分排序水平线 + horizontal_lines.sort(key=lambda x: x[4], reverse=True) - if len(horizontal_lines) > 1: - # 按y坐标排序 (从大到小，底部的线排在前面) - horizontal_lines.sort(key=lambda x: x[1], reverse=True) - - # 提取最底部和次底部的线段 - bottom_line = horizontal_lines[0] - - # 检查是否有明显的上下边缘 - y_coords = [line[1] for line in horizontal_lines] - y_diffs = [y_coords[i] - y_coords[i+1] for i in range(len(y_coords)-1)] - - if len(y_diffs) > 0 and max(y_diffs) > height * 0.05: # 如果有明显的高度差 - # 找到高度差最大的位置 - split_idx = y_diffs.index(max(y_diffs)) - - # 分别获取下边缘线和上边缘线 - bottom_line = horizontal_lines[0] # 最底部的线 - top_line = horizontal_lines[split_idx+1] # 上边缘线 - - # 检查上下两条线是否平行 - 计算斜率差异 - bottom_slope = bottom_line[2] - top_slope = top_line[2] - slope_diff = abs(bottom_slope - top_slope) - - # 计算两线的交点（如果存在） - bottom_x1, bottom_y1, bottom_x2, bottom_y2 = bottom_line[0] - top_x1, top_y1, top_x2, top_y2 = top_line[0] - - # 根据斜率差异判断是否平行 - # 如果斜率差异很小，认为基本平行 - if slope_diff > 0.05: # 斜率差异超过阈值 - # 计算两条线延长后的交点 - # 线段方程: y = mx + b - # 计算两条线的截距b - bottom_b = bottom_y1 - bottom_slope * bottom_x1 - top_b = top_y1 - top_slope * top_x1 - - # 检查交点是否在图像范围内或者附近 - # 求解 y = m1*x + b1 = m2*x + b2 - if abs(bottom_slope - top_slope) > 1e-6: # 避免除以接近0的值 - intersection_x = (top_b - bottom_b) / (bottom_slope - top_slope) - intersection_y = bottom_slope * intersection_x + bottom_b - - # 判断交点是否在图像宽度的2倍范围内 - if -width <= intersection_x <= width * 2: - is_truncated = True - if observe: - debug(f"检测到上下边缘线不平行，交点: ({intersection_x:.1f}, {intersection_y:.1f})", "分析") - debug("判断下方线被截断", "分析") - - # 显示上下边缘线及其延长线和交点 - intersect_img = img.copy() - # 画原始线段 - cv2.line(intersect_img, (bottom_x1, bottom_y1), (bottom_x2, bottom_y2), (0, 255, 0), 2) - cv2.line(intersect_img, (top_x1, top_y1), (top_x2, top_y2), (255, 0, 255), 2) - - # 延长线段以显示交点 - ext_left_x = max(0, int(intersection_x - width/2)) - ext_right_x = min(width-1, int(intersection_x + width/2)) - - # 计算延长线上的点 - bottom_ext_left_y = int(bottom_slope * ext_left_x + bottom_b) - bottom_ext_right_y = int(bottom_slope * ext_right_x + bottom_b) - top_ext_left_y = int(top_slope * ext_left_x + top_b) - top_ext_right_y = int(top_slope * ext_right_x + top_b) - - # 绘制延长线 - cv2.line(intersect_img, (ext_left_x, bottom_ext_left_y), - (ext_right_x, bottom_ext_right_y), (0, 255, 0), 1, cv2.LINE_DASHED) - cv2.line(intersect_img, (ext_left_x, top_ext_left_y), - (ext_right_x, top_ext_right_y), (255, 0, 255), 1, cv2.LINE_DASHED) - - # 标记交点 - if 0 <= intersection_x < width and 0 <= intersection_y < height: - cv2.circle(intersect_img, (int(intersection_x), int(intersection_y)), - 10, (0, 0, 255), -1) - - cv2.imshow("上下边缘线交点分析", intersect_img) - cv2.waitKey(delay) - - # 如果检测到下方线被截断，使用上边缘线来估计实际的下边缘线 - if is_truncated and top_line is not None: - if observe: - debug("使用上边缘估计真实的下边缘", "处理") - - # 获取赛道平均宽度（可以是预先测量的固定值，或根据未截断部分测量） - # 这里假设赛道宽度是固定的，可以根据实际情况调整 - track_width_pixels = height * 0.15 # 假设赛道宽度是图像高度的15% - - # 计算一个修正后的底部线段，方向与上边缘线平行，但位置下移 - corrected_bottom_slope = top_slope # 使用上边缘的斜率 - - # 计算上边缘线的方程: y = mx + b - top_b = top_y1 - top_slope * top_x1 - - # 计算修正后的下边缘线的截距，使其下移track_width_pixels距离 - # 由于是在图像坐标系，y轴向下，所以是加法 - corrected_bottom_b = top_b + track_width_pixels - - # 计算修正后的下边缘线的两个端点 - corrected_bottom_x1 = left_bound - corrected_bottom_y1 = int(corrected_bottom_slope * corrected_bottom_x1 + corrected_bottom_b) - corrected_bottom_x2 = right_bound - corrected_bottom_y2 = int(corrected_bottom_slope * corrected_bottom_x2 + corrected_bottom_b) - - # 创建修正后的底部线段 - corrected_bottom_line = ( - [corrected_bottom_x1, corrected_bottom_y1, corrected_bottom_x2, corrected_bottom_y2], - (corrected_bottom_y1 + corrected_bottom_y2) / 2, # mid_y - corrected_bottom_slope, - np.sqrt((corrected_bottom_x2-corrected_bottom_x1)**2 + (corrected_bottom_y2-corrected_bottom_y1)**2) # length - ) - - if observe: - # 显示修正后的线段 - corrected_img = img.copy() - # 原始线段 - cv2.line(corrected_img, (bottom_x1, bottom_y1), (bottom_x2, bottom_y2), (0, 255, 0), 2) - cv2.line(corrected_img, (top_x1, top_y1), (top_x2, top_y2), (255, 0, 255), 2) - # 修正后的线段 - cv2.line(corrected_img, (corrected_bottom_x1, corrected_bottom_y1), - (corrected_bottom_x2, corrected_bottom_y2), (0, 0, 255), 2) - - cv2.putText(corrected_img, "截断的底边", (bottom_x1, bottom_y1 - 10), - cv2.FONT_HERSHEY_SIMPLEX, 0.5, (0, 255, 0), 2) - cv2.putText(corrected_img, "上边缘", (top_x1, top_y1 - 10), - cv2.FONT_HERSHEY_SIMPLEX, 0.5, (255, 0, 255), 2) - cv2.putText(corrected_img, "修正后的底边", (corrected_bottom_x1, corrected_bottom_y1 - 10), - cv2.FONT_HERSHEY_SIMPLEX, 0.5, (0, 0, 255), 2) - - cv2.imshow("修正后的边缘线", corrected_img) - cv2.waitKey(delay) - - # 使用修正后的底部线作为选定的线 - bottom_line = corrected_bottom_line - else: - # 只有一条水平线 - bottom_line = horizontal_lines[0] - - # 使用底部线段作为最终选择 - selected_line = bottom_line[0] - selected_slope = bottom_line[2] + # 取质量最高的线段作为最终选择 + selected_line = horizontal_lines[0][0] + selected_slope = horizontal_lines[0][2] + selected_score = horizontal_lines[0][4] # 提取线段端点 x1, y1, x2, y2 = selected_line @@ -338,6 +224,59 @@ def detect_horizontal_track_edge(image, observe=False, delay=1000, save_log=True else: bottom_edge_point = (x2, y2) + # 如果得分过低，可能是错误识别，尝试使用边缘点拟合 + if selected_score < 0.4 and len(bottom_points) >= 5: + if observe: + debug(f"线段质量得分过低: {selected_score:.2f}，尝试使用边缘点拟合", "处理") + + # 筛选下半部分的点 + valid_bottom_points = [p for p in bottom_points if p[1] >= valid_y_range[0]] + + if len(valid_bottom_points) >= 5: + # 使用RANSAC拟合直线以去除异常值 + x_points = np.array([p[0] for p in valid_bottom_points]).reshape(-1, 1) + y_points = np.array([p[1] for p in valid_bottom_points]) + + ransac = linear_model.RANSACRegressor(residual_threshold=5.0) + ransac.fit(x_points, y_points) + + # 获取拟合参数 + fitted_slope = ransac.estimator_.coef_[0] + intercept = ransac.estimator_.intercept_ + + # 检查斜率是否在合理范围内 + if abs(fitted_slope) < max_slope: + # 计算拟合线的inliers比例 + inlier_mask = ransac.inlier_mask_ + inlier_ratio = sum(inlier_mask) / len(inlier_mask) + + # 如果有足够的内点，使用拟合的直线 + if inlier_ratio > 0.5: + # 使用拟合的直线参数计算线段端点 + x1 = left_bound + y1 = int(fitted_slope * x1 + intercept) + x2 = right_bound + y2 = int(fitted_slope * x2 + intercept) + + selected_slope = fitted_slope + selected_line = [x1, y1, x2, y2] + + # 重新计算边缘点 + if y1 > y2: + bottom_edge_point = (x1, y1) + else: + bottom_edge_point = (x2, y2) + + if observe: + debug(f"使用拟合直线，斜率: {fitted_slope:.4f}, 内点比例: {inlier_ratio:.2f}", "处理") + fitted_line_img = img.copy() + cv2.line(fitted_line_img, (x1, y1), (x2, y2), (0, 255, 255), 2) + for i, point in enumerate(valid_bottom_points): + color = (0, 255, 0) if inlier_mask[i] else (0, 0, 255) + cv2.circle(fitted_line_img, point, 3, color, -1) + cv2.imshow("拟合线和内点", fitted_line_img) + cv2.waitKey(delay) + # 获取线上的更多点 selected_points = [] step = 5 # 每5个像素取一个点 @@ -346,22 +285,34 @@ def detect_horizontal_track_edge(image, observe=False, delay=1000, save_log=True if top_bound <= y <= bottom_bound: selected_points.append((x, y)) - if observe: - debug(f"步骤7: 找到边缘点 {bottom_edge_point}", "检测") - edge_img = img.copy() - # 画线 - cv2.line(edge_img, (x1, y1), (x2, y2), (0, 255, 0), 2) - # 绘制所有点 - for point in selected_points: - cv2.circle(edge_img, point, 3, (255, 0, 0), -1) - # 标记边缘点 - cv2.circle(edge_img, bottom_edge_point, 10, (0, 0, 255), -1) - cv2.imshow("选定的横向线和边缘点", edge_img) - cv2.waitKey(delay) + # 对结果进行合理性检查 + valid_result = True + reason = "" + + # 检查边缘点是否在有效范围内 + if not (valid_y_range[0] <= bottom_edge_point[1] <= valid_y_range[1]): + valid_result = False + reason += "边缘点y坐标超出有效范围; " + + # 检查线段长度 + line_length = np.sqrt((x2-x1)**2 + (y2-y1)**2) + if line_length < min_line_length: + valid_result = False + reason += "线段长度不足; " + + # 检查是否有足够的点 + if len(selected_points) < 5: + valid_result = False + reason += "选定点数量不足; " # 计算这个点到中线的距离 distance_to_center = bottom_edge_point[0] - center_x + # 检查到中心的距离是否合理 + if abs(distance_to_center) > width * 0.8: + valid_result = False + reason += "到中心距离过大; " + # 计算中线与检测到的横向线的交点 # 横向线方程: y = slope * (x - x1) + y1 # 中线方程: x = center_x @@ -370,14 +321,24 @@ def detect_horizontal_track_edge(image, observe=False, delay=1000, save_log=True intersection_y = selected_slope * (center_x - x1) + y1 intersection_point = (int(intersection_x), int(intersection_y)) + # 检查交点的y坐标是否在有效范围内 + if not (valid_y_range[0] <= intersection_y <= valid_y_range[1]): + valid_result = False + reason += "交点y坐标超出有效范围; " + # 计算交点到图像底部的距离（以像素为单位） distance_to_bottom = height - intersection_y + # 如果结果无效，可能需要返回失败 + if not valid_result and observe: + warning(f"检测结果不合理: {reason}", "警告") + result_img = None if observe or save_log: slope_img = img.copy() # 画出检测到的线 - cv2.line(slope_img, (x1, y1), (x2, y2), (0, 255, 0), 2) + line_color = (0, 255, 0) if valid_result else (0, 0, 255) + cv2.line(slope_img, (x1, y1), (x2, y2), line_color, 2) # 标记边缘点 cv2.circle(slope_img, bottom_edge_point, 10, (0, 0, 255), -1) # 画出中线 @@ -389,16 +350,19 @@ def detect_horizontal_track_edge(image, observe=False, delay=1000, save_log=True cv2.line(slope_img, intersection_point, (intersection_x, height), (255, 255, 0), 2) cv2.putText(slope_img, f"Slope: {selected_slope:.4f}", (10, 30), - cv2.FONT_HERSHEY_SIMPLEX, 1, (0, 255, 0), 2) + cv2.FONT_HERSHEY_SIMPLEX, 1, line_color, 2) cv2.putText(slope_img, f"Distance to center: {distance_to_center}px", (10, 70), - cv2.FONT_HERSHEY_SIMPLEX, 1, (0, 255, 0), 2) + cv2.FONT_HERSHEY_SIMPLEX, 1, line_color, 2) cv2.putText(slope_img, f"Distance to bottom: {distance_to_bottom:.1f}px", (10, 110), - cv2.FONT_HERSHEY_SIMPLEX, 1, (0, 255, 0), 2) + cv2.FONT_HERSHEY_SIMPLEX, 1, line_color, 2) cv2.putText(slope_img, f"中线交点: ({intersection_point[0]}, {intersection_point[1]})", (10, 150), - cv2.FONT_HERSHEY_SIMPLEX, 1, (0, 255, 0), 2) - if is_truncated: - cv2.putText(slope_img, "下方线被截断(已修正)", (10, 190), - cv2.FONT_HERSHEY_SIMPLEX, 1, (0, 0, 255), 2) + cv2.FONT_HERSHEY_SIMPLEX, 1, line_color, 2) + cv2.putText(slope_img, f"质量得分: {selected_score:.2f}", (10, 190), + cv2.FONT_HERSHEY_SIMPLEX, 1, line_color, 2) + + if not valid_result: + cv2.putText(slope_img, f"警告: {reason}", (10, 230), + cv2.FONT_HERSHEY_SIMPLEX, 0.8, (0, 0, 255), 2) if observe: debug("显示边缘斜率和中线交点", "显示") @@ -424,10 +388,16 @@ def detect_horizontal_track_edge(image, observe=False, delay=1000, save_log=True "slope": selected_slope, "distance_to_bottom": distance_to_bottom, "intersection_point": intersection_point, - "is_truncated": is_truncated + "score": selected_score, + "valid": valid_result, + "reason": reason if not valid_result else "" } info(f"横向边缘检测结果: {log_info}", "日志") + # 如果结果无效，可能需要返回失败 + if not valid_result: + return None, None + # 创建边缘信息字典 edge_info = { "x": bottom_edge_point[0], @@ -438,7 +408,7 @@ def detect_horizontal_track_edge(image, observe=False, delay=1000, save_log=True "points_count": len(selected_points), # 该组中点的数量 "intersection_point": intersection_point, # 中线与横向线的交点 "distance_to_bottom": distance_to_bottom, # 交点到图像底部的距离 - "is_truncated": is_truncated, # 下方线是否被截断并修正 + "score": selected_score, # 线段质量得分 # "points": selected_points # 添加选定的点组 }