OpenCV第六篇：阈值函数

阳光穿透心脏的1/2处 2024-04-08 11:34 57阅读 0赞

阈值的作用是根据设定的值处理图像的灰度值，比如灰度大于某个数值像素点保留。通过阈值以及有关算法可以实现从图像中抓取特定的图形，比如去除背景等。实例图片：

#### ![在这里插入图片描述][watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L1NHY2hp_size_16_color_FFFFFF_t_70] 1.普通阈值函数：threshold(像素矩阵,起始阈值,最大值,算法类型)-->retval, threshold ####

import cv2
    
    if __name__ == '__main__':
        img = cv2.imread('1.png')
        retval, threshold = cv2.threshold(img, 12, 255, cv2.THRESH_BINARY)
        cv2.imshow('original', img)
        cv2.imshow('threshold', threshold)
        cv2.waitKey(0)
        cv2.destroyAllWindows()

![583c488c6db84d00ac703eb88b63d2b5.png][]

使用图片的灰度图试试：

import cv2
    
    if __name__ == '__main__':
        img = cv2.imread('1.png')
        # 灰度图
        gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
        retval, threshold = cv2.threshold(gray, 12, 255, cv2.THRESH_BINARY)
        cv2.imshow('original', img)
        cv2.imshow('threshold', threshold)
        cv2.waitKey(0)
        cv2.destroyAllWindows()

![93de28482d894dbb960338c61a897e07.png][]

#### 2.自适应阈值函数：adaptivthreshold() ####

import cv2
    
    if __name__ == '__main__':
        img = cv2.imread('1.png')
        # 灰度图
        gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
        threshold = cv2.adaptiveThreshold(gray, 255, cv2.ADAPTIVE_THRESH_GAUSSIAN_C, cv2.THRESH_BINARY, 115, 1)
        cv2.imshow('original', img)
        cv2.imshow('threshold', threshold)
        cv2.waitKey(0)
        cv2.destroyAllWindows()

![bc04394087714ea892a591b3c7060728.png][]

[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L1NHY2hp_size_16_color_FFFFFF_t_70]: https://image.dandelioncloud.cn/pgy_files/images/2024/04/08/45ab82680f5744fca68e2b5148ccd728.png
[583c488c6db84d00ac703eb88b63d2b5.png]: https://image.dandelioncloud.cn/pgy_files/images/2024/04/08/316406d349fd4cc4afc4c8170717d9c2.png
[93de28482d894dbb960338c61a897e07.png]: https://image.dandelioncloud.cn/pgy_files/images/2024/04/08/462efeae85f84b5ca259fb5a7d863e7a.png
[bc04394087714ea892a591b3c7060728.png]: https://image.dandelioncloud.cn/pgy_files/images/2024/04/08/53db57885fde4df292ee2b42d6511996.png