<div id="readability-page-1" class="page"><div>

 warpPerspective 함수를 사용하여 퍼스펙티브 변환(Perspective Transformation)을 구현합니다.





 퍼스펙티브 변환에서 원본 이미지의 모든 직선은 출력 이미지에서 직선으로 유지됩니다. 





 퍼스펙티브 변환 행렬을 찾으려면 입력 이미지의 4점과 대응하는 출력 이미지의 4점이 필요합니다.





 getPerspectiveTransform 함수를 사용하면 대응하는 4점 쌍에 대한 변환 행렬을 구할 수 있습니다.





 warpPerspective 함수를 사용하여 변환을 실행합니다.





 마지막 업데이트 – 2018. 10. 4









 다음 순서대로 책의 모서리를 클릭합니다.




<figure><img data-lazy-fallback="1" data-attachment-id="5743" data-permalink="https://webnautes.blog/5742/word-image-5742-1/" data-orig-file="https://i0.wp.com/webnautes.blog/wp-content/uploads/2023/08/word-image-5742-1.png?fit=678%2C789&amp;ssl=1" data-orig-size="678,789" data-comments-opened="1" data-image-meta="{&quot;aperture&quot;:&quot;0&quot;,&quot;credit&quot;:&quot;&quot;,&quot;camera&quot;:&quot;&quot;,&quot;caption&quot;:&quot;&quot;,&quot;created_timestamp&quot;:&quot;0&quot;,&quot;copyright&quot;:&quot;&quot;,&quot;focal_length&quot;:&quot;0&quot;,&quot;iso&quot;:&quot;0&quot;,&quot;shutter_speed&quot;:&quot;0&quot;,&quot;title&quot;:&quot;&quot;,&quot;orientation&quot;:&quot;0&quot;}" data-image-title="word-image-5742-1" data-image-description="" data-image-caption="" data-medium-file="https://i0.wp.com/webnautes.blog/wp-content/uploads/2023/08/word-image-5742-1.png?fit=258%2C300&amp;ssl=1" data-large-file="https://i0.wp.com/webnautes.blog/wp-content/uploads/2023/08/word-image-5742-1.png?fit=678%2C789&amp;ssl=1" decoding="async" fetchpriority="high" width="678" height="789" src="https://i0.wp.com/webnautes.blog/wp-content/uploads/2023/08/word-image-5742-1.png?resize=678%2C789&amp;ssl=1" alt="" sizes="(max-width: 678px) 100vw, 678px" data-recalc-dims="1" data-old-src="https://i0.wp.com/webnautes.blog/wp-content/uploads/2023/08/word-image-5742-1.png?resize=678%2C789&amp;is-pending-load=1#038;ssl=1" data-lazy-srcset="https://i0.wp.com/webnautes.blog/wp-content/uploads/2023/08/word-image-5742-1.png?w=678&amp;ssl=1 678w, https://i0.wp.com/webnautes.blog/wp-content/uploads/2023/08/word-image-5742-1.png?resize=258%2C300&amp;ssl=1 258w"></figure>








 클릭한 책 모서리에 빨간점이 찍히게 됩니다.




<figure><img data-lazy-fallback="1" data-attachment-id="5744" data-permalink="https://webnautes.blog/5742/word-image-5742-2/" data-orig-file="https://i0.wp.com/webnautes.blog/wp-content/uploads/2023/08/word-image-5742-2.png?fit=678%2C789&amp;ssl=1" data-orig-size="678,789" data-comments-opened="1" data-image-meta="{&quot;aperture&quot;:&quot;0&quot;,&quot;credit&quot;:&quot;&quot;,&quot;camera&quot;:&quot;&quot;,&quot;caption&quot;:&quot;&quot;,&quot;created_timestamp&quot;:&quot;0&quot;,&quot;copyright&quot;:&quot;&quot;,&quot;focal_length&quot;:&quot;0&quot;,&quot;iso&quot;:&quot;0&quot;,&quot;shutter_speed&quot;:&quot;0&quot;,&quot;title&quot;:&quot;&quot;,&quot;orientation&quot;:&quot;0&quot;}" data-image-title="word-image-5742-2" data-image-description="" data-image-caption="" data-medium-file="https://i0.wp.com/webnautes.blog/wp-content/uploads/2023/08/word-image-5742-2.png?fit=258%2C300&amp;ssl=1" data-large-file="https://i0.wp.com/webnautes.blog/wp-content/uploads/2023/08/word-image-5742-2.png?fit=678%2C789&amp;ssl=1" decoding="async" width="678" height="789" src="https://i0.wp.com/webnautes.blog/wp-content/uploads/2023/08/word-image-5742-2.png?resize=678%2C789&amp;ssl=1" alt="" sizes="(max-width: 678px) 100vw, 678px" data-recalc-dims="1" data-old-src="https://i0.wp.com/webnautes.blog/wp-content/uploads/2023/08/word-image-5742-2.png?resize=678%2C789&amp;is-pending-load=1#038;ssl=1" data-lazy-srcset="https://i0.wp.com/webnautes.blog/wp-content/uploads/2023/08/word-image-5742-2.png?w=678&amp;ssl=1 678w, https://i0.wp.com/webnautes.blog/wp-content/uploads/2023/08/word-image-5742-2.png?resize=258%2C300&amp;ssl=1 258w"></figure>








 스페이스바를 누르면 정면에서 바라본 책의 이미지로 변환된 이미지가 보입니다. 




<figure><img data-lazy-fallback="1" data-attachment-id="5745" data-permalink="https://webnautes.blog/5742/word-image-5742-3/" data-orig-file="https://i0.wp.com/webnautes.blog/wp-content/uploads/2023/08/word-image-5742-3.png?fit=678%2C789&amp;ssl=1" data-orig-size="678,789" data-comments-opened="1" data-image-meta="{&quot;aperture&quot;:&quot;0&quot;,&quot;credit&quot;:&quot;&quot;,&quot;camera&quot;:&quot;&quot;,&quot;caption&quot;:&quot;&quot;,&quot;created_timestamp&quot;:&quot;0&quot;,&quot;copyright&quot;:&quot;&quot;,&quot;focal_length&quot;:&quot;0&quot;,&quot;iso&quot;:&quot;0&quot;,&quot;shutter_speed&quot;:&quot;0&quot;,&quot;title&quot;:&quot;&quot;,&quot;orientation&quot;:&quot;0&quot;}" data-image-title="word-image-5742-3" data-image-description="" data-image-caption="" data-medium-file="https://i0.wp.com/webnautes.blog/wp-content/uploads/2023/08/word-image-5742-3.png?fit=258%2C300&amp;ssl=1" data-large-file="https://i0.wp.com/webnautes.blog/wp-content/uploads/2023/08/word-image-5742-3.png?fit=678%2C789&amp;ssl=1" decoding="async" width="678" height="789" src="https://i0.wp.com/webnautes.blog/wp-content/uploads/2023/08/word-image-5742-3.png?resize=678%2C789&amp;ssl=1" alt="" sizes="(max-width: 678px) 100vw, 678px" data-recalc-dims="1" data-old-src="https://i0.wp.com/webnautes.blog/wp-content/uploads/2023/08/word-image-5742-3.png?resize=678%2C789&amp;is-pending-load=1#038;ssl=1" data-lazy-srcset="https://i0.wp.com/webnautes.blog/wp-content/uploads/2023/08/word-image-5742-3.png?w=678&amp;ssl=1 678w, https://i0.wp.com/webnautes.blog/wp-content/uploads/2023/08/word-image-5742-3.png?resize=258%2C300&amp;ssl=1 258w"></figure>








 테스트에 사용한 원본 이미지와 전체 코드입니다.




<figure><img data-lazy-fallback="1" data-attachment-id="5746" data-permalink="https://webnautes.blog/5742/word-image-5742-4/" data-orig-file="https://i0.wp.com/webnautes.blog/wp-content/uploads/2023/08/word-image-5742-4.png?fit=675%2C756&amp;ssl=1" data-orig-size="675,756" data-comments-opened="1" data-image-meta="{&quot;aperture&quot;:&quot;0&quot;,&quot;credit&quot;:&quot;&quot;,&quot;camera&quot;:&quot;&quot;,&quot;caption&quot;:&quot;&quot;,&quot;created_timestamp&quot;:&quot;0&quot;,&quot;copyright&quot;:&quot;&quot;,&quot;focal_length&quot;:&quot;0&quot;,&quot;iso&quot;:&quot;0&quot;,&quot;shutter_speed&quot;:&quot;0&quot;,&quot;title&quot;:&quot;&quot;,&quot;orientation&quot;:&quot;0&quot;}" data-image-title="word-image-5742-4" data-image-description="" data-image-caption="" data-medium-file="https://i0.wp.com/webnautes.blog/wp-content/uploads/2023/08/word-image-5742-4.png?fit=268%2C300&amp;ssl=1" data-large-file="https://i0.wp.com/webnautes.blog/wp-content/uploads/2023/08/word-image-5742-4.png?fit=675%2C756&amp;ssl=1" decoding="async" loading="lazy" width="675" height="756" src="https://i0.wp.com/webnautes.blog/wp-content/uploads/2023/08/word-image-5742-4.png?resize=675%2C756&amp;ssl=1" alt="" sizes="(max-width: 675px) 100vw, 675px" data-recalc-dims="1" data-old-src="https://i0.wp.com/webnautes.blog/wp-content/uploads/2023/08/word-image-5742-4.png?resize=675%2C756&amp;is-pending-load=1#038;ssl=1" data-lazy-srcset="https://i0.wp.com/webnautes.blog/wp-content/uploads/2023/08/word-image-5742-4.png?w=675&amp;ssl=1 675w, https://i0.wp.com/webnautes.blog/wp-content/uploads/2023/08/word-image-5742-4.png?resize=268%2C300&amp;ssl=1 268w"></figure>







<pre><code>import numpy as np
import cv2


point_list = []
count = 0

def mouse_callback(event, x, y, flags, param):
 global point_list, count, img_original


 # 마우스 왼쪽 버튼 누를 때마다 좌표를 리스트에 저장
 if event == cv2.EVENT_LBUTTONDOWN:
 print("(%d, %d)" % (x, y))
 point_list.append((x, y))

 print(point_list)
 cv2.circle(img_original, (x, y), 3, (0, 0, 255), -1)



cv2.namedWindow('original')
cv2.setMouseCallback('original', mouse_callback)

# 원본 이미지
img_original = cv2.imread('test3.png')


while(True):

 cv2.imshow("original", img_original)


 height, width = img_original.shape[:2]


 if cv2.waitKey(1)&amp;0xFF == 32: # spacebar를 누르면 루프에서 빠져나옵니다.
 break


# 좌표 순서 - 상단왼쪽 끝, 상단오른쪽 끝, 하단왼쪽 끝, 하단오른쪽 끝
pts1 = np.float32([list(point_list[0]),list(point_list[1]),list(point_list[2]),list(point_list[3])])
pts2 = np.float32([[0,0],[width,0],[0,height],[width,height]])

print(pts1)
print(pts2)

M = cv2.getPerspectiveTransform(pts1,pts2)

img_result = cv2.warpPerspective(img_original, M, (width,height))


cv2.imshow("result1", img_result)
cv2.waitKey(0)
cv2.destroyAllWindows()</code></pre>





</div></div>

멈춤보단 천천히라도

webnautes

warpPerspective 함수를 사용하여 퍼스펙티브 변환(Perspective Transformation)을 구현합니다. 퍼스펙티브 변환에서 원본 이미지의 모든 직선은 출력 이미지에서 직선으로 유지됩니다. 퍼스펙티브 변환 행렬을 찾으려면 입력 이미지의 4점과 대응하는 출력 이미지의 4점이 필요합니다. getPerspectiveTransform 함수를 사용하면 대응하는 4점 쌍에 대한 변환 행렬을 구할 수 있습니다. warpPerspective 함수를 사용하여 변환을 실행합니다. 마지막 업데이트 – 2018. 10. 4 다음 순서대로 책의… 더 보기 »OpenCV Python 강좌 – Perspective Transformation
The post OpenCV Python 강좌 – Perspective Transformation first appeared on 멈춤보단 천천히라도.

OpenCV Python 강좌 – Perspective Transformation

댓글