114 lines
3.1 KiB
Python
114 lines
3.1 KiB
Python
# 利用OpenCV实现两张2D图像的3D重构。
|
|
|
|
import cv2
|
|
import numpy as np
|
|
import plotly.express as px
|
|
import plotly.io as pio
|
|
|
|
# 设置plotly在浏览器中显示
|
|
pio.renderers.default = "browser"
|
|
|
|
# 读取图像
|
|
img1 = cv2.imread(r"10\templeRing\templeR0003.png")
|
|
img2 = cv2.imread(r"10\templeRing\templeR0005.png")
|
|
|
|
# 定义相机内参
|
|
fx = 0.25 * 1520.4
|
|
fy = 0.25 * 1525.9
|
|
cx = 302.32
|
|
cy = 246.87
|
|
|
|
K = np.array([[fx, 0, cx], [0, fy, cy], [0, 0, 1]])
|
|
|
|
# 定义第一个相机的外参矩阵
|
|
R1 = np.array(
|
|
[
|
|
[-0.01625331773280620100, 0.98386957700862299000, -0.17814736905031653000],
|
|
[0.97668439268305030000, -0.02252259937820530100, -0.21349550254417543000],
|
|
[-0.21406407160478280000, -0.17746376518636725000, -0.96056399333613396000],
|
|
]
|
|
)
|
|
t1 = np.array([[-0.0283090812583], [-0.0366442193256], [0.529139415773]])
|
|
|
|
# 定义第二个相机的外参矩阵
|
|
R2 = np.array(
|
|
[
|
|
[-0.05235090589954815400, 0.98479784589115438000, -0.16562785206491965000],
|
|
[0.88539496349116698000, -0.03093875352325572600, -0.46380874523331522000],
|
|
[-0.46188217250287084000, -0.17092687400923678000, -0.87031538103463302000],
|
|
]
|
|
)
|
|
t2 = np.array([[-0.0269600886818], [-0.0469344855587], [0.53860946783]])
|
|
|
|
# 转换图像为灰度图
|
|
gray1 = cv2.cvtColor(img1, cv2.COLOR_BGR2GRAY)
|
|
gray2 = cv2.cvtColor(img2, cv2.COLOR_BGR2GRAY)
|
|
|
|
# 创建SIFT特征检测器
|
|
sift = cv2.SIFT_create()
|
|
|
|
# 检测特征点和计算描述符
|
|
kp1, des1 = sift.detectAndCompute(gray1, None)
|
|
kp2, des2 = sift.detectAndCompute(gray2, None)
|
|
|
|
# 创建BFMatcher对象
|
|
bf = cv2.BFMatcher(cv2.NORM_L2, crossCheck=True)
|
|
|
|
# 进行特征匹配
|
|
matches = bf.match(des1, des2)
|
|
|
|
# 按距离排序
|
|
matches = sorted(matches, key=lambda x: x.distance)
|
|
|
|
# 选择前N个最佳匹配点（可以调整这个数值）
|
|
N_BEST_MATCHES = 100
|
|
matches = matches[:N_BEST_MATCHES]
|
|
|
|
# 提取匹配点的坐标
|
|
pts1 = np.float32([kp1[m.queryIdx].pt for m in matches]).reshape(-1, 1, 2)
|
|
pts2 = np.float32([kp2[m.trainIdx].pt for m in matches]).reshape(-1, 1, 2)
|
|
|
|
# 计算投影矩阵
|
|
P1 = K @ np.hstack((R1, t1))
|
|
P2 = K @ np.hstack((R2, t2))
|
|
|
|
# 三角测量得到3D点
|
|
points_4D = cv2.triangulatePoints(P1, P2, pts1, pts2)
|
|
points_3D = points_4D / points_4D[3]
|
|
points_3D = points_3D[:3, :].T
|
|
|
|
# 可选：移除一些可能的异常点
|
|
# 这里使用简单的距离阈值进行过滤
|
|
distances = np.sqrt(np.sum(points_3D**2, axis=1))
|
|
mask = distances < np.percentile(distances, 95) # 移除最远的5%的点
|
|
points_3D = points_3D[mask]
|
|
|
|
# 使用plotly进行3D可视化
|
|
fig = px.scatter_3d(
|
|
x=points_3D[:, 0], y=points_3D[:, 1], z=points_3D[:, 2], title="3D Reconstruction"
|
|
)
|
|
|
|
# 调整点的大小和视角
|
|
fig.update_traces(marker=dict(size=1))
|
|
fig.update_layout(scene=dict(aspectmode="data")) # 保持真实比例
|
|
|
|
# 显示结果
|
|
fig.show()
|
|
|
|
# 保存结果
|
|
fig.write_html("10/templeR.html")
|
|
|
|
# 可选：显示匹配结果
|
|
img_matches = cv2.drawMatches(
|
|
img1,
|
|
kp1,
|
|
img2,
|
|
kp2,
|
|
matches,
|
|
None,
|
|
flags=cv2.DrawMatchesFlags_NOT_DRAW_SINGLE_POINTS,
|
|
)
|
|
cv2.imshow("Matches", img_matches)
|
|
cv2.waitKey(0)
|
|
cv2.destroyAllWindows()
|