Skip to content
Merged
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
183 changes: 74 additions & 109 deletions src/lane_visualize/main.py
Original file line number Diff line number Diff line change
@@ -1,7 +1,8 @@
import cv2
import numpy as np
import matplotlib.pyplot as plt

import sys
import os

class LaneDetector:
def __init__(self):
Expand All @@ -11,202 +12,166 @@ def __init__(self):
self.threshold = 15
self.min_line_length = 40
self.max_line_gap = 20

# 车道区域顶点(需要根据摄像头位置调整)
self.vertices = None

def region_of_interest(self, img):
"""
定义感兴趣区域(ROI),只处理图像中可能包含车道的部分
"""
"""定义感兴趣区域(ROI)"""
mask = np.zeros_like(img)

# 如果顶点未定义,使用默认值
if self.vertices is None:
height, width = img.shape
# 定义梯形区域,底部为图像底部,顶部为图像中心附近
# 调整梯形区域以适应大多数行车记录仪视角
self.vertices = np.array([[
(width * 0.1, height),
(width * 0.45, height * 0.6),
(width * 0.55, height * 0.6),
(width * 0.9, height)
(width * 0.1, height), # 左下
(width * 0.45, height * 0.6), # 左上
(width * 0.55, height * 0.6), # 右上
(width * 0.9, height) # 右下
]], dtype=np.int32)

# 填充多边形区域
cv2.fillPoly(mask, self.vertices, 255)

# 应用掩码
masked_img = cv2.bitwise_and(img, mask)
return masked_img

def detect_edges(self, img):
"""
使用Canny边缘检测算法检测图像中的边缘
"""
# 转换为灰度图
"""Canny边缘检测"""
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)

# 高斯模糊以减少噪声
blur = cv2.GaussianBlur(gray, (5, 5), 0)

# Canny边缘检测
edges = cv2.Canny(blur, 50, 150)

return edges

def detect_lines(self, edges):
"""
使用霍夫变换检测直线
"""
"""霍夫变换检测直线"""
lines = cv2.HoughLinesP(edges, self.rho, self.theta, self.threshold,
np.array([]), minLineLength=self.min_line_length,
maxLineGap=self.max_line_gap)
return lines

def average_slope_intercept(self, lines):
"""
将检测到的线段按左右车道线分组,并计算平均斜率和截距
"""
left_lines = [] # 左车道线
right_lines = [] # 右车道线
"""计算左右车道线的平均斜率和截距"""
left_lines = []
right_lines = []

if lines is not None:
for line in lines:
x1, y1, x2, y2 = line[0]

# 计算斜率
if x2 - x1 == 0: # 避免除以零
continue
if x2 - x1 == 0: continue
slope = (y2 - y1) / (x2 - x1)

# 过滤掉水平线(斜率接近0)和垂直线(斜率非常大)
# 过滤不合理的斜率 (太水平或太垂直)
if abs(slope) < 0.5 or abs(slope) > 2:
continue

# 计算截距
intercept = y1 - slope * x1

# 根据斜率正负区分左右车道线
if slope < 0: # 左车道线斜率为负
# 图像坐标系中,y轴向下,所以斜率为负是左车道,正为右车道
if slope < 0:
left_lines.append((slope, intercept))
else: # 右车道线斜率为正
else:
right_lines.append((slope, intercept))

# 计算平均斜率和截距
left_avg = np.average(left_lines, axis=0) if left_lines else None
right_avg = np.average(right_lines, axis=0) if right_lines else None

return left_avg, right_avg

def make_line_points(self, avg_line, y_min, y_max):
"""
根据斜率和截距生成直线的端点
"""
"""生成绘制用的坐标点"""
if avg_line is None:
return None

slope, intercept = avg_line

# 计算直线与y_min和y_max的交点
x_min = int((y_min - intercept) / slope)
x_max = int((y_max - intercept) / slope)

return [(x_min, y_min), (x_max, y_max)]
# 防止除以0
if slope == 0:
return None

try:
x_min = int((y_min - intercept) / slope)
x_max = int((y_max - intercept) / slope)
return [(x_min, y_min), (x_max, y_max)]
except OverflowError:
return None

def draw_lane(self, img, left_line, right_line):
"""
在图像上绘制车道区域
"""
# 创建空白图像
"""绘制半透明车道区域"""
lane_img = np.zeros_like(img)


# 确保两条线都检测到了才画多边形
if left_line is not None and right_line is not None:
# 提取点坐标
left_pts = np.array([left_line[0], left_line[1]], dtype=np.int32)
right_pts = np.array([right_line[0], right_line[1]], dtype=np.int32)

# 创建车道多边形
# 创建多边形顶点
pts = np.vstack([left_pts, np.flipud(right_pts)])
cv2.fillPoly(lane_img, [pts], (0, 255, 0)) # 绿色填充

# 无论是否形成区域,都尝试画线
if left_line is not None:
cv2.line(lane_img, left_line[0], left_line[1], (255, 0, 0), 10) # 蓝色线
if right_line is not None:
cv2.line(lane_img, right_line[0], right_line[1], (255, 0, 0), 10) # 蓝色线

# 填充车道区域
cv2.fillPoly(lane_img, [pts], (0, 255, 0))

# 绘制车道线
cv2.line(lane_img, left_pts[0], left_pts[1], (255, 0, 0), 5)
cv2.line(lane_img, right_pts[0], right_pts[1], (255, 0, 0), 5)

# 将车道图像叠加到原图上
result = cv2.addWeighted(img, 0.8, lane_img, 0.4, 0)
return result

def process_frame(self, frame):
"""
处理单帧图像,检测车道并绘制结果
"""
# 检测边缘
if frame is None: return None
edges = self.detect_edges(frame)

# 提取感兴趣区域
roi_edges = self.region_of_interest(edges)

# 检测直线
lines = self.detect_lines(roi_edges)

# 获取图像尺寸

height, width = frame.shape[:2]

# 计算平均车道线
left_avg, right_avg = self.average_slope_intercept(lines)

# 生成车道线端点
y_min = int(height * 0.6) # 车道线上端点
y_max = height # 车道线下端点

y_min = int(height * 0.65) # 稍作调整,不要画太远
y_max = height

left_line = self.make_line_points(left_avg, y_min, y_max)
right_line = self.make_line_points(right_avg, y_min, y_max)

# 绘制车道

result = self.draw_lane(frame, left_line, right_line)

return result


def main():
# 创建车道检测器
detector = LaneDetector()

# 打开摄像头(0为默认摄像头)
cap = cv2.VideoCapture(0)

# 或者打开视频文件
# cap = cv2.VideoCapture('road_video.mp4')

# 逻辑:如果有命令行参数,读取视频;否则读取摄像头
input_source = 0 # 默认摄像头

if len(sys.argv) > 1:
input_path = sys.argv[1]
if os.path.exists(input_path):
input_source = input_path
print(f"正在打开视频文件: {input_path}")
else:
print(f"错误: 找不到文件 {input_path}")
return
else:
print("未提供视频路径,正在打开默认摄像头...")

cap = cv2.VideoCapture(input_source)

if not cap.isOpened():
print("无法打开摄像头")
print("错误: 无法打开视频源")
return

print("按 'q' 键退出程序")

while True:
# 读取帧
ret, frame = cap.read()

if not ret:
print("无法读取帧,退出...")
print("视频播放结束或无法读取帧")
break

# 处理帧
processed_frame = detector.process_frame(frame)

# 显示结果
cv2.imshow('Lane Detection', processed_frame)
if processed_frame is not None:
cv2.imshow('Lane Detection System', processed_frame)
else:
cv2.imshow('Lane Detection System', frame)

# 按'q'退出
if cv2.waitKey(1) & 0xFF == ord('q'):
# 按 'q' 退出,设置 25ms 延时(约 40fps)
if cv2.waitKey(25) & 0xFF == ord('q'):
break

# 释放资源
cap.release()
cv2.destroyAllWindows()


if __name__ == "__main__":
main()
Loading