高动态星点提取方法

python版本

• 3.6.0 - 3.8.0

已有功能

• 静态星点仿真
• 动态星点仿真
• 估计星点的旋转中心（平行时为无穷远处）
• 图像增强、星点修复
• 姿态解算、重投影

当前工作

• 解决重投影误差较大的问题

依赖库

• numpy、maplotlib、opencv、sklearn、scipy、PIL

目录对应说明

• estimate: 高动态星图运动方向估计程序
• generator: 动静态星图仿真生成程序
• extract: 星点提取程序
• identifier: 星图识别、姿态解算、重投影程序
• params: 一些参数或者星库文件
• tools: 一些工具程序
• graph: 星图文件夹
• temp: 临时文件夹，舍不得删的放这里

项目使用示例

1. 静态仿真星图生成

import matplotlib.pyplot as plt
import generator.generate_starImg as ggs

if __name__ == "__main__":
    
    # Attitude: [yaw, pitch, roll]
    att = [20, 40, 60]

    # Generate
    retImg = ggs.genStatic(att)

    # Show
    plt.figure()
    plt.imshow(retImg, cmap='gray', vmin=0, vmax=255)
    plt.show()

2. 动态仿真星图生成

import matplotlib.pyplot as plt
import generator.generate_starImg as ggs

if __name__ == "__main__":
    
    # Attitude: (yaw, pitch, roll)
    att = [20, 40, 60]

    # Angle Velocity: (yaw, pitch, roll)
    dps = [5, 5, 5]

    # Generate
    retImg = ggs.genDynamic(att, dps, 100)

    # Show
    plt.figure()
    plt.imshow(retImg, cmap='gray', vmin=0, vmax=255)
    plt.show()

3. 旋转中心估计

import cv2
from cv2 import cv2

import estimate.est as ae

if __name__ == "__main__":

    # Read source image
    filename = r'./graph/test.png'
    src = cv2.imread(filename, 0)
    
    # Estimate the rotation-center
    src = cv2.blur(src, (3, 3))
    rot_center = ae.Direction_estimate(src)
    print('Rcenter:', rot_center)

4. 星点提取、质心定位

import cv2
from cv2 import cv2

import estimate.est as ae
import extract.extract as ex

if __name__ == "__main__":

    # Read source image
    filename = r'./graph/test.png'
    src = cv2.imread(filename, 0)
    
    # Estimate the rotation-center
    src = cv2.blur(src, (3, 3))
    rot_center = ae.Direction_estimate(src)
    print('Rcenter:', rot_center)
    
    # Star extraction and centroid location
    retImg, centers, cnt = ex.extract(src.copy(), rot_center)
    centers = centers[:cnt]
    print('Stars:', cnt)

5. 星图识别、姿态解算、重投影

import cv2
from cv2 import cv2

import estimate.est as ae
import extract.extract as ex
import identifier.pyramid as ip

if __name__ == "__main__":

    # Read source image
    filename = r'./graph/test.png'
    src = cv2.imread(filename, 0)
    
    # Estimate the rotation-center
    src = cv2.blur(src, (3, 3))
    rot_center = ae.Direction_estimate(src)
    print('Rcenter:', rot_center)
    # rot_center = [999999999, 999999999]

    # Star extraction and centroid location
    retImg, centers, cnt = ex.extract(src.copy(), rot_center)
    centers = centers[:cnt]
    print('Stars:', cnt)

    # Star identification
    att = ip.identify(centers)
    print('Attitude:', att)
    
    # Re-projection
    retImg, iden = ip.reProjection(retImg, att, centers)
    print('Identified:', iden)
    plt.figure()
    # retImg = retImg.astype(np.int32)
    # retImg[np.where(retImg > 255)] = 255
    # plt.imshow(retImg, cmap='gray', vmin=0, vmax=255)
    plt.imshow(retImg)
    if rot_center[0] != 999999999:
        plt.scatter(rot_center[0], rot_center[1], s = 1)
    plt.show()