利用GRB颜色模型过滤条件进行火灾检测

本是古典何须时尚 2024-04-19 10:16 50阅读 0赞

火灾检测的相关介绍参考我另一篇博客：[火灾检测方法总结][Link 1]  
      本文RGB过滤条件参考论文：[Fire detection using statistical color model in video sequences][]，提取效果比一般的RGB颜色过滤提取效果好些，我进行了测试，提取的区域更加饱和，代码参考博客：[Opencv + python火焰分割加速版][Opencv _ python]进行改写，其中的GRB过滤条件为：  
                                                            ![在这里插入图片描述][watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3pob3U0NDExNzgx_size_16_color_FFFFFF_t_70]

import cv2 as cv
    import numpy as np
    
    def contrast_brightness_demo(image, c, b):  #其中c为对比度，b为每个像素加上的值（调节亮度）
        blank = np.zeros(image.shape, image.dtype)   #创建一张与原图像大小及通道数都相同的黑色图像
        dst = cv.addWeighted(image, c, blank, 1-c, b) #c为加权值，b为每个像素所加的像素值
        ret, dst = cv.threshold(dst, 25, 255, cv.THRESH_BINARY)
        return dst
    
    
    capture = cv.VideoCapture("/home/yasin/house5.mp4")
    
    while(True):
        ret, frame = capture.read()
        cv.imshow("frame", frame)
        B = frame[:, :, 0]
        G = frame[:, :, 1]
        R = frame[:, :, 2]
        R_mean = np.mean(R)
        val1 = G /(R + 1)
        val2 = B /(R + 1)
        val3 = B /(G + 1) 
        fireImg = np.array(np.where(R > R_mean, np.where(R >= G, np.where(G >= B, np.where(val1 >= 0.25, np.where(val1 <=0.65, np.where(val2 >= 0.05,np.where(val2 <=0.45,np.where(val3 >=0.2 ,np.where(val3 <= 0.6, 255, 0), 0), 0), 0), 0), 0), 0), 0), 0))
        
        gray_fireImg = np.zeros([fireImg.shape[0], fireImg.shape[1], 1], np.uint8)
        gray_fireImg[:, :, 0] = fireImg
        gray_fireImg = cv.GaussianBlur(gray_fireImg, (7, 7), 0)
        gray_fireImg = contrast_brightness_demo(gray_fireImg, 5.0, 25)
        kernel = cv.getStructuringElement(cv.MORPH_RECT, (5, 5))
        gray_fireImg = cv.morphologyEx(gray_fireImg, cv.MORPH_CLOSE, kernel)
        dst = cv.bitwise_and(frame, frame, mask=gray_fireImg)
        cv.namedWindow("fire",0)
        cv.imshow("fire", dst)
        cv.namedWindow("gray_fireImg",0)
        cv.imshow("gray_fireImg", gray_fireImg)
        c = cv.waitKey(40)
        if c == 27:
            break

原始图：  
                                  ![在这里插入图片描述][watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3pob3U0NDExNzgx_size_16_color_FFFFFF_t_70 1]

测试结果：  
                                ![在这里插入图片描述][watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3pob3U0NDExNzgx_size_16_color_FFFFFF_t_70 2]

[Link 1]: https://blog.csdn.net/zhou4411781/article/details/100095274
[Fire detection using statistical color model in video sequences]: https://www.sciencedirect.com/science/article/pii/S1047320306000927
[Opencv _ python]: https://blog.csdn.net/weixin_41987641/article/details/81812823
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3pob3U0NDExNzgx_size_16_color_FFFFFF_t_70]: https://image.dandelioncloud.cn/pgy_files/images/2024/04/18/f84e84d1e9d8419a8f128cdbf16d7acd.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3pob3U0NDExNzgx_size_16_color_FFFFFF_t_70 1]: https://image.dandelioncloud.cn/pgy_files/images/2024/04/18/f362cf541b304b17b75378a833499486.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3pob3U0NDExNzgx_size_16_color_FFFFFF_t_70 2]: https://image.dandelioncloud.cn/pgy_files/images/2024/04/18/318789ae29aa4f1c9fb129bfe4f6f1e5.png