如何提取并可视化多条2D红色线条的高密度交叉区域？

我用不同颜色的2D线条连接多个数据点，但线条数量多的时候，很难区分大部分线条重叠或交叉的区域。以下是我的绘图脚本：

ax.scatter(rcv[:,0], rcv[:,1], c='b', marker='v',s=190,label= "End")
ax.scatter(src[:,0], src[:,1], c='r', marker='*',s=120,label= "Start")

if status=="High":
    x_values = [x1, x2]
    y_values = [y1, y2]

    plt.plot(x_values, y_values, 'k-', linewidth=0.5,alpha=0.5)
    fig.canvas.draw()
else:
    x_values = [x1, x2]
    y_values = [y1, y2]

    plt.plot(x_values, y_values, 'r-', linewidth=0.5,alpha=0.5)
    fig.canvas.draw()

脚本中绘制了若干红色和黑色线条，我希望提取红色线条交叉密度较高的区域（支持用户自定义交叉数量阈值），恳请提供可行的解决方案建议。

方法1：网格密度统计法 #

核心思路是把绘图区域拆分成均匀网格，统计每个网格内经过的红色线条数量，再根据阈值筛选高密度区域。

实现代码示例： #

import numpy as np

# 第一步：提前收集所有红色线条的线段数据
red_lines = []  # 在绘图的else分支中，把每条红线的(x1,y1,x2,y2)追加到这个列表

# 第二步：定义网格范围与分辨率
x_min, x_max = ax.get_xlim()
y_min, y_max = ax.get_ylim()
grid_size = 50  # 网格数量，可根据需求调整
x_grid = np.linspace(x_min, x_max, grid_size)
y_grid = np.linspace(y_min, y_max, grid_size)
density_grid = np.zeros((grid_size-1, grid_size-1))

# 第三步：判断线段是否与网格相交的辅助函数
def line_intersects_grid(x1, y1, x2, y2, grid_x_left, grid_x_right, grid_y_bottom, grid_y_top):
    # 先判断线段端点是否在网格内
    in_grid = (grid_x_left <= x1 <= grid_x_right and grid_y_bottom <= y1 <= grid_y_top) or \
              (grid_x_left <= x2 <= grid_x_right and grid_y_bottom <= y2 <= grid_y_top)
    if in_grid:
        return True
    # 简化判断：检查线段是否穿过网格边界（如需更精确可引入专业线段-矩形相交算法）
    return False

# 第四步：统计每个网格的线条覆盖数
for x1, y1, x2, y2 in red_lines:
    for i in range(grid_size-1):
        for j in range(grid_size-1):
            grid_x_left = x_grid[i]
            grid_x_right = x_grid[i+1]
            grid_y_bottom = y_grid[j]
            grid_y_top = y_grid[j+1]
            if line_intersects_grid(x1, y1, x2, y2, grid_x_left, grid_x_right, grid_y_bottom, grid_y_top):
                density_grid[j, i] += 1

# 第五步：自定义阈值并可视化高密度区域
threshold = 10  # 用户可根据需求调整
high_density_mask = density_grid >= threshold

for i in range(grid_size-1):
    for j in range(grid_size-1):
        if high_density_mask[j, i]:
            rect = plt.Rectangle((x_grid[i], y_grid[j]), x_grid[i+1]-x_grid[i], y_grid[j+1]-y_grid[j], 
                                facecolor='yellow', alpha=0.3)
            ax.add_patch(rect)

方法2：基于像素的热力图法 #

利用画布像素统计红色线条的覆盖次数，将其转换为热力图后筛选高密度区域。

实现代码示例： #

# 第一步：创建空白画布单独绘制红色线条
fig_density, ax_density = plt.subplots(figsize=fig.get_size_inches(), dpi=fig.dpi)
ax_density.set_xlim(ax.get_xlim())
ax_density.set_ylim(ax.get_ylim())
ax_density.set_facecolor('black')  # 黑色背景，白色线条便于统计

# 第二步：绘制所有红色线条
for x1, y1, x2, y2 in red_lines:
    ax_density.plot([x1, x2], [y1, y2], 'w-', linewidth=0.5)

# 第三步：提取画布像素数据
fig_density.canvas.draw()
width, height = fig_density.canvas.get_width_height()
pixel_data = np.frombuffer(fig_density.canvas.tostring_rgb(), dtype=np.uint8).reshape(height, width, 3)
white_channel = pixel_data[:, :, 0]  # 白色线条区域值为255，覆盖次数越多值越高
plt.close(fig_density)

# 第四步：自定义阈值筛选高密度像素
threshold = 200  # 值越高，要求线条覆盖次数越多
high_density_pixels = white_channel >= threshold

# 第五步：将高密度区域叠加到原图
ax.imshow(high_density_pixels, extent=[x_min, x_max, y_min, y_max], origin='lower', cmap='YlOrRd', alpha=0.3)

方法3：交点密度统计法 #

直接计算所有红色线条间的交点，统计交点在各区域的分布密度，筛选超过阈值的区域。

实现代码示例： #

# 第一步：计算两条线段交点的辅助函数
def get_line_intersection(x1, y1, x2, y2, x3, y3, x4, y4):
    denom = (x1 - x2)*(y3 - y4) - (y1 - y2)*(x3 - x4)
    if denom == 0:
        return None  # 线段平行或重合
    t_num = (x1 - x3)*(y3 - y4) - (y1 - y3)*(x3 - x4)
    u_num = (x1 - x3)*(y1 - y2) - (y1 - y3)*(x1 - x2)
    t = t_num / denom
    u = -u_num / denom
    if 0 <= t <= 1 and 0 <= u <= 1:
        x = x1 + t*(x2 - x1)
        y = y1 + t*(y2 - y1)
        return (x, y)
    return None

# 第二步：计算所有红色线条间的交点
intersections = []
n_lines = len(red_lines)
for i in range(n_lines):
    x1, y1, x2, y2 = red_lines[i]
    for j in range(i+1, n_lines):
        x3, y3, x4, y4 = red_lines[j]
        pt = get_line_intersection(x1, y1, x2, y2, x3, y3, x4, y4)
        if pt is not None:
            intersections.append(pt)

# 第三步：统计交点的网格密度
if intersections:
    xs, ys = zip(*intersections)
    x_min, x_max = min(xs), max(xs)
    y_min, y_max = min(ys), max(ys)
    grid_size = 30
    x_grid = np.linspace(x_min, x_max, grid_size)
    y_grid = np.linspace(y_min, y_max, grid_size)
    density, _, _ = np.histogram2d(xs, ys, bins=[x_grid, y_grid])
    
    # 第四步：自定义阈值并可视化
    threshold = 5  # 交点数量阈值，用户可调整
    high_density_mask = density >= threshold
    
    for i in range(grid_size-1):
        for j in range(grid_size-1):
            if high_density_mask[j, i]:
                rect = plt.Rectangle((x_grid[i], y_grid[j]), x_grid[i+1]-x_grid[i], y_grid[j+1]-y_grid[j], 
                                    facecolor='orange', alpha=0.4)
                ax.add_patch(rect)

内容的提问来源于stack exchange，提问作者Aqeel