import cv2 import numpy as np import os import datetime from utils.log_helper import LogHelper, get_logger, section, info, debug, warning, error, success, timing def detect_dual_track_lines(image, observe=False, delay=1000, save_log=True, stone_path_mode=False): """ 检测左右两条平行的黄色轨道线，优化后能够处理石板路上的黄线 参数: image: 输入图像，可以是文件路径或者已加载的图像数组 observe: 是否输出中间状态信息和可视化结果，默认为False delay: 展示每个步骤的等待时间(毫秒) save_log: 是否保存日志和图像 stone_path_mode: 石板路模式，针对石板路上的黄线进行特殊处理 返回: tuple: (中心线信息, 左轨迹线信息, 右轨迹线信息) """ # 如果输入是字符串（文件路径），则加载图像 if isinstance(image, str): img = cv2.imread(image) else: img = image.copy() if img is None: error("无法加载图像", "失败") return None, None, None # 获取图像尺寸 height, width = img.shape[:2] # 计算图像中间区域的范围 center_x = width // 2 # 转换到HSV颜色空间以便更容易提取黄色 hsv = cv2.cvtColor(img, cv2.COLOR_BGR2HSV) # 根据是否为石板路模式选择不同的参数 if stone_path_mode: # 石板路上的黄线通常对比度更低，需要更宽松的颜色范围 lower_yellow = np.array([10, 60, 60]) # 更宽松的黄色下限 upper_yellow = np.array([40, 255, 255]) # 更宽松的黄色上限 else: # 标准黄色的HSV范围 lower_yellow = np.array([15, 80, 80]) upper_yellow = np.array([35, 255, 255]) # 创建黄色的掩码 mask = cv2.inRange(hsv, lower_yellow, upper_yellow) # 应用对比度增强 if stone_path_mode: # 在处理掩码前先对原图像进行预处理 # 对原图进行自适应直方图均衡化增强对比度 lab = cv2.cvtColor(img, cv2.COLOR_BGR2LAB) l, a, b = cv2.split(lab) clahe = cv2.createCLAHE(clipLimit=3.0, tileGridSize=(8, 8)) l = clahe.apply(l) lab = cv2.merge((l, a, b)) enhanced_img = cv2.cvtColor(lab, cv2.COLOR_LAB2BGR) # 用增强后的图像重新检测黄色 enhanced_hsv = cv2.cvtColor(enhanced_img, cv2.COLOR_BGR2HSV) enhanced_mask = cv2.inRange(enhanced_hsv, lower_yellow, upper_yellow) # 组合原始掩码和增强掩码 mask = cv2.bitwise_or(mask, enhanced_mask) if observe: debug("增强对比度和颜色检测", "处理") cv2.imshow("增强对比度", enhanced_img) cv2.imshow("增强后的黄色掩码", enhanced_mask) cv2.waitKey(delay) # 形态学操作以改善掩码 if stone_path_mode: # 石板路上的线条可能更细更断续，使用更大的膨胀核和更多的迭代次数 kernel = np.ones((7, 7), np.uint8) mask = cv2.dilate(mask, kernel, iterations=2) mask = cv2.erode(mask, np.ones((3, 3), np.uint8), iterations=1) else: kernel = np.ones((5, 5), np.uint8) mask = cv2.dilate(mask, kernel, iterations=1) mask = cv2.erode(mask, np.ones((3, 3), np.uint8), iterations=1) if observe: debug("步骤1: 创建黄色掩码", "处理") cv2.imshow("黄色掩码", mask) cv2.waitKey(delay) # 裁剪底部区域重点关注近处的黄线 bottom_roi_height = int(height * 0.4) # 关注图像底部40%区域 bottom_roi = mask[height-bottom_roi_height:, :] if observe: debug("步骤1.5: 底部区域掩码", "处理") cv2.imshow("底部区域掩码", bottom_roi) cv2.waitKey(delay) # 边缘检测 - 针对石板路调整参数 if stone_path_mode: edges = cv2.Canny(mask, 30, 120, apertureSize=3) # 降低阈值以捕捉更弱的边缘 else: edges = cv2.Canny(mask, 50, 150, apertureSize=3) if observe: debug("步骤2: 边缘检测", "处理") cv2.imshow("边缘检测", edges) cv2.waitKey(delay) # 霍夫变换检测直线 - 根据是否为石板路调整参数 if stone_path_mode: # 石板路上的线段可能更短更断续，使用更宽松的参数 lines = cv2.HoughLinesP(edges, 1, np.pi/180, threshold=20, minLineLength=width*0.03, maxLineGap=50) else: lines = cv2.HoughLinesP(edges, 1, np.pi/180, threshold=25, minLineLength=width*0.05, maxLineGap=40) if lines is None or len(lines) == 0: error("未检测到直线", "失败") return None, None, None if observe: debug(f"步骤3: 检测到 {len(lines)} 条直线", "处理") lines_img = img.copy() for line in lines: x1, y1, x2, y2 = line[0] cv2.line(lines_img, (x1, y1), (x2, y2), (0, 255, 0), 2) cv2.imshow("检测到的直线", lines_img) cv2.waitKey(delay) # 筛选近似垂直的线 vertical_lines = [] for line in lines: x1, y1, x2, y2 = line[0] # 优先选择图像底部的线 if y1 < height * 0.5 and y2 < height * 0.5: continue # 忽略上半部分的线 # 计算斜率 (避免除零错误) if abs(x2 - x1) < 5: # 几乎垂直的线 slope = 100 # 设置一个较大的值表示接近垂直 else: slope = (y2 - y1) / (x2 - x1) # 根据是否为石板路调整斜率阈值 min_slope_threshold = 0.5 if stone_path_mode else 0.75 # 筛选接近垂直的线 (斜率较大)，但允许更多倾斜度 if abs(slope) > min_slope_threshold: line_length = np.sqrt((x2-x1)**2 + (y2-y1)**2) # 计算线的中点x坐标 mid_x = (x1 + x2) / 2 # 计算线的中点y坐标 mid_y = (y1 + y2) / 2 # 保存线段、其坐标、斜率和长度 vertical_lines.append((line[0], mid_x, mid_y, slope, line_length)) if len(vertical_lines) < 2: # 石板路模式下，尝试放宽斜率条件重新筛选线段 if stone_path_mode: vertical_lines = [] for line in lines: x1, y1, x2, y2 = line[0] # 仍然优先选择图像底部的线 if y1 < height * 0.6 and y2 < height * 0.6: continue # 忽略上部分的线 # 计算斜率 (避免除零错误) if abs(x2 - x1) < 5: # 几乎垂直的线 slope = 100 else: slope = (y2 - y1) / (x2 - x1) # 使用更宽松的斜率阈值 if abs(slope) > 0.3: line_length = np.sqrt((x2-x1)**2 + (y2-y1)**2) mid_x = (x1 + x2) / 2 mid_y = (y1 + y2) / 2 vertical_lines.append((line[0], mid_x, mid_y, slope, line_length)) if len(vertical_lines) < 2: error("未检测到足够的垂直线", "失败") return None, None, None if observe: debug(f"步骤4: 找到 {len(vertical_lines)} 条垂直线", "处理") v_lines_img = img.copy() for line_info in vertical_lines: line, _, _, slope, _ = line_info x1, y1, x2, y2 = line cv2.line(v_lines_img, (x1, y1), (x2, y2), (0, 255, 255), 2) # 显示斜率 cv2.putText(v_lines_img, f"{slope:.2f}", ((x1+x2)//2, (y1+y2)//2), cv2.FONT_HERSHEY_SIMPLEX, 0.5, (255, 0, 0), 1) cv2.imshow("垂直线", v_lines_img) cv2.waitKey(delay) # 优先选择更接近图像底部的线 - 根据y坐标均值排序 vertical_lines.sort(key=lambda x: x[2], reverse=True) # 按mid_y从大到小排序 # 石板路模式下，可能需要处理断续的线段或合并相近的线段 if stone_path_mode and len(vertical_lines) >= 3: # 尝试合并相近的线段 merged_lines = [] processed = [False] * len(vertical_lines) for i in range(len(vertical_lines)): if processed[i]: continue current_line = vertical_lines[i] _, current_mid_x, _, current_slope, current_length = current_line # 查找相近的线段 similar_lines = [current_line] processed[i] = True for j in range(i+1, len(vertical_lines)): if processed[j]: continue candidate_line = vertical_lines[j] _, candidate_mid_x, _, candidate_slope, candidate_length = candidate_line # 如果x坐标接近且斜率相似，认为是同一条线的不同部分 x_diff = abs(current_mid_x - candidate_mid_x) slope_diff = abs(current_slope - candidate_slope) if x_diff < width * 0.05 and slope_diff < 0.3: similar_lines.append(candidate_line) processed[j] = True # 如果找到多条相近的线，合并它们 if len(similar_lines) > 1: # 按线长度加权合并 total_weight = sum(line[4] for line in similar_lines) merged_x1 = sum(line[0][0] * line[4] for line in similar_lines) / total_weight merged_y1 = sum(line[0][1] * line[4] for line in similar_lines) / total_weight merged_x2 = sum(line[0][2] * line[4] for line in similar_lines) / total_weight merged_y2 = sum(line[0][3] * line[4] for line in similar_lines) / total_weight merged_line = (np.array([int(merged_x1), int(merged_y1), int(merged_x2), int(merged_y2)]), (merged_x1 + merged_x2) / 2, (merged_y1 + merged_y2) / 2, (merged_y2 - merged_y1) / (merged_x2 - merged_x1 + 1e-6), np.sqrt((merged_x2-merged_x1)**2 + (merged_y2-merged_y1)**2)) merged_lines.append(merged_line) else: merged_lines.append(current_line) vertical_lines = merged_lines if observe: debug(f"步骤4.5: 合并后找到 {len(vertical_lines)} 条垂直线", "处理") merged_img = img.copy() for line_info in vertical_lines: line, _, _, slope, _ = line_info x1, y1, x2, y2 = line cv2.line(merged_img, (int(x1), int(y1)), (int(x2), int(y2)), (255, 0, 255), 2) # 显示斜率 cv2.putText(merged_img, f"{slope:.2f}", (int((x1+x2)//2), int((y1+y2)//2)), cv2.FONT_HERSHEY_SIMPLEX, 0.5, (255, 0, 0), 1) cv2.imshow("合并后的垂直线", merged_img) cv2.waitKey(delay) # 按x坐标将线分为左右两组 left_lines = [line for line in vertical_lines if line[1] < center_x] right_lines = [line for line in vertical_lines if line[1] > center_x] # 如果任一侧没有检测到线，则放宽左右两侧线的分组条件 if not left_lines or not right_lines: # 按x坐标排序所有垂直线 vertical_lines.sort(key=lambda x: x[1]) # 如果有至少两条线，将最左侧的线作为左轨迹线，最右侧的线作为右轨迹线 if len(vertical_lines) >= 2: left_lines = [vertical_lines[0]] right_lines = [vertical_lines[-1]] else: error("左侧或右侧未检测到轨迹线", "失败") return None, None, None # 从左右两组中各选择一条最佳的线 # 优先选择同时满足：1. 更靠近底部 2. 足够长 3. 更接近中心的线 def score_line(line_info, is_left): _, mid_x, mid_y, slope, length = line_info # y越大（越靠近底部）分数越高 y_score = mid_y / height # 线越长分数越高 length_score = min(1.0, length / (height * 0.3)) # 石板路模式下，调整预期的线位置 if stone_path_mode: # 使用更宽的轨道宽度估计 expected_x = center_x * 0.25 if is_left else center_x * 1.75 else: expected_x = center_x * 0.3 if is_left else center_x * 1.7 x_score = 1.0 - min(1.0, abs(mid_x - expected_x) / (center_x * 0.5)) # 综合评分，石板路模式下更重视线段长度 if stone_path_mode: return y_score * 0.4 + length_score * 0.4 + x_score * 0.2 else: return y_score * 0.5 + length_score * 0.3 + x_score * 0.2 # 对左右线组进行评分并排序 left_lines = sorted(left_lines, key=lambda line: score_line(line, True), reverse=True) right_lines = sorted(right_lines, key=lambda line: score_line(line, False), reverse=True) left_line = left_lines[0] right_line = right_lines[0] # 获取两条线的坐标 left_x1, left_y1, left_x2, left_y2 = left_line[0] right_x1, right_y1, right_x2, right_y2 = right_line[0] # 确保线段的顺序是从上到下 if left_y1 > left_y2: left_x1, left_x2 = left_x2, left_x1 left_y1, left_y2 = left_y2, left_y1 if right_y1 > right_y2: right_x1, right_x2 = right_x2, right_x1 right_y1, right_y2 = right_y2, right_y1 # 计算中心线 center_line_x1 = (left_x1 + right_x1) // 2 center_line_y1 = (left_y1 + right_y1) // 2 center_line_x2 = (left_x2 + right_x2) // 2 center_line_y2 = (left_y2 + right_y2) // 2 # 计算中心线的斜率 if abs(center_line_x2 - center_line_x1) < 5: center_slope = 100 # 几乎垂直 else: center_slope = (center_line_y2 - center_line_y1) / (center_line_x2 - center_line_x1) # 计算中心线延伸到图像底部的点 if abs(center_slope) < 0.01: # 几乎垂直 bottom_x = center_line_x1 else: bottom_x = int(center_line_x1 + (height - center_line_y1) / center_slope) center_point = (bottom_x, height) # 计算中心线与图像中心线的偏差 deviation = bottom_x - center_x result_img = None if observe or save_log: result_img = img.copy() # 绘制左右轨迹线 cv2.line(result_img, (int(left_x1), int(left_y1)), (int(left_x2), int(left_y2)), (255, 0, 0), 2) cv2.line(result_img, (int(right_x1), int(right_y1)), (int(right_x2), int(right_y2)), (0, 0, 255), 2) # 绘制中心线 cv2.line(result_img, (center_line_x1, center_line_y1), (center_line_x2, center_line_y2), (0, 255, 0), 2) cv2.line(result_img, (center_line_x2, center_line_y2), center_point, (0, 255, 0), 2) # 绘制图像中心线 cv2.line(result_img, (center_x, 0), (center_x, height), (0, 0, 255), 1) # 标记中心点 cv2.circle(result_img, center_point, 10, (255, 0, 255), -1) # 显示偏差信息 cv2.putText(result_img, f"Deviation: {deviation}px", (10, 30), cv2.FONT_HERSHEY_SIMPLEX, 1, (0, 255, 0), 2) # 显示是否为石板路模式 if stone_path_mode: cv2.putText(result_img, "Stone Path Mode", (10, 60), cv2.FONT_HERSHEY_SIMPLEX, 1, (0, 255, 255), 2) if observe: cv2.imshow("轨迹线检测结果", result_img) cv2.waitKey(delay) # 保存日志图像 if save_log and result_img is not None: timestamp = datetime.datetime.now().strftime("%Y%m%d_%H%M%S_%f") log_dir = "logs/image" os.makedirs(log_dir, exist_ok=True) img_path = os.path.join(log_dir, f"dual_track_{timestamp}.jpg") cv2.imwrite(img_path, result_img) info(f"保存双轨迹线检测结果图像到: {img_path}", "日志") # 保存文本日志信息 log_info = { "timestamp": timestamp, "center_point": center_point, "deviation": deviation, "left_track_mid_x": left_line[1], "right_track_mid_x": right_line[1], "track_width": right_line[1] - left_line[1], "center_slope": center_slope, "stone_path_mode": stone_path_mode } info(f"双轨迹线检测结果: {log_info}", "日志") # 创建左右轨迹线和中心线信息 left_track_info = { "line": left_line[0], "slope": left_line[3], "x_mid": left_line[1] } right_track_info = { "line": right_line[0], "slope": right_line[3], "x_mid": right_line[1] } center_info = { "point": center_point, "deviation": deviation, "slope": center_slope, "is_vertical": abs(center_slope) > 5.0, # 判断是否接近垂直 "track_width": right_line[1] - left_line[1], # 两轨迹线之间的距离 "stone_path_mode": stone_path_mode # 记录是否使用了石板路模式 } return center_info, left_track_info, right_track_info def auto_detect_dual_track_lines(image, observe=False, delay=1000, save_log=True): """ 自动检测双轨迹线，先尝试标准模式，如果失败则尝试石板路模式 参数: image: 输入图像，可以是文件路径或者已加载的图像数组 observe: 是否输出中间状态信息和可视化结果，默认为False delay: 展示每个步骤的等待时间(毫秒) save_log: 是否保存日志和图像 返回: tuple: (中心线信息, 左轨迹线信息, 右轨迹线信息) """ # 先尝试标准模式 result = detect_dual_track_lines(image, observe, delay, save_log, stone_path_mode=False) # 如果标准模式失败，尝试石板路模式 if result[0] is None: info("标准模式检测失败，尝试石板路模式", "自动检测") result = detect_dual_track_lines(image, observe, delay, save_log, stone_path_mode=True) return result