惊鸿若梦一书生
人工智能方向硕士和博士。已发表SCI(二区)论文1篇,EI论文4篇,著作1部,并多次获得国家级证书,主持申报并结项三项科研基金项目,累计经费逾50W。中国自动化协会会员,CSDN专栏作家,博文累计阅读量逾120W。
展开
-
Pandas实战100例 | 案例 98: 分组和聚合
在数据分析中,经常需要根据某些列的值进行分组,并对每个组进行聚合计算。Pandas 提供了用于分组和聚合的方法。原创 2024-01-19 08:50:45 · 86 阅读 · 0 评论 -
OpenCV-Day-023:实战 | 给图片加文字
# -*- coding: utf-8 -*-"""@ModuleName:pic_add_name@Function: @Author: H2017824@Time: 2020/6/8 19:53"""import pandas as pdimport cv2import ostable_order = 5src_path = 'std_{}.xlsx'.format(table_order)df_std_1 = pd.read_excel(src_path)name_list原创 2020-10-27 10:06:52 · 136 阅读 · 0 评论 -
OpenCV-Day-022:实战 | 之水印提取和去除水印
图片下载:代码import cv2import numpy as nppath = "1111.png"# 读取原始数据img = cv2.imread(path)cv2.imshow('img', img)print('img.shape:', img.shape)# height, width = img.shape[0:2]# 色彩阈值化处理,选择感兴趣的区域thresh = cv2.inRange(img, np.array([135, 187, 155]), np.ar原创 2020-08-19 09:55:43 · 1889 阅读 · 0 评论 -
OpenCV-Day-021:实战 | 识别图片中的二维码(python+opencv+pyzbar)
代码import pyzbar.pyzbar as pyzbarimport cv2 as cvsrc = cv.imread('./multi.jpg', cv.IMREAD_COLOR)cv.namedWindow('input', cv.WINDOW_NORMAL)cv.imshow('input', src)barcodes = pyzbar.decode(src)fo...原创 2019-06-08 10:03:34 · 1506 阅读 · 2 评论 -
OpenCV-Day-020:图像直方图反向投影
代码实验效果解释重点:RGB色彩空间对一些纯色的区分度不够,HSV, YCaCb这些色彩空间对它们的区分度会比较好!图像直方图反向投影是通过构建指定模板图像的二维直方图空间与目标的二维直方图空间,进行直方图数据归一化之后, 进行比率操作,对所有得到非零数值,生成查找表对原图像进行像素映射之后,再进行图像模糊输出的结果。直方图反向投影流程:计算直方图计算比率RLUT查找表...原创 2019-06-08 09:25:33 · 347 阅读 · 2 评论 -
OpenCV-Day-019:图像直方图比较
代码实验效果解释图像直方图比较,就是计算两幅图像的直方图数据,比较两组数据的相似性,从而得到两幅图像之间的相似程度,直方图比较在早期的CBIR中是应用很常见的技术手段,通常会结合边缘处理、词袋等技术一起使用。OpenCV中直方图比较的API很简单compareHist(hist1, hist2, method)-常见比较方法有相关性卡方交叉巴氏其中相关性计算和巴氏距离...原创 2019-06-08 09:11:53 · 272 阅读 · 3 评论 -
OpenCV-Day-018:图像直方图均衡化
代码import cv2 as cvimport matplotlib.pyplot as pltimport numpy as npdef custom_hist(gray): h, w = gray.shape hist = np.zeros([256], dtype=np.int32) for row in range(h): for co...原创 2019-06-07 08:07:04 · 257 阅读 · 0 评论 -
OpenCV-Day-017:图像直方图
代码import cv2 as cvimport matplotlib.pyplot as pltimport numpy as np# 自己编写程序实现灰度图像直方图的统计def custom_hist(gray): h, w = gray.shape hist = np.zeros([256], dtype=np.int32) for row in ran...原创 2019-06-06 21:43:30 · 222 阅读 · 0 评论 -
OpenCV-Day-016:图像ROI和ROI操作
代码import cv2 as cvimport numpy as npsrc = cv.imread("./apple.jpg", cv.IMREAD_COLOR)cv.namedWindow("input", cv.WINDOW_AUTOSIZE)cv.imshow("input", src)h, w = src.shape[:2]print(h, w)# 获取ROIcy ...原创 2019-06-05 09:49:14 · 656 阅读 · 0 评论 -
OpenCV-Day-015:几何形状绘制
代码import cv2 as cvimport numpy as npsrc = np.zeros((512, 512, 3), dtype=np.uint8)# 线宽为2cv.rectangle(src, (100, 100), (300, 300), (255, 0, 0), 2, cv.LINE_8, 0)cv.circle(src, (255, 255), 50, (0, ...原创 2019-06-04 23:36:24 · 295 阅读 · 0 评论 -
OpenCV-Day-014:图像插值
代码import cv2 as cvsrc = cv.imread('./0.jpg', cv.IMREAD_COLOR)h, w = src.shape[:2]print(h, w)dst = cv.resize(src, (w * 2, h * 2), fx=0, fy=0, interpolation=cv.INTER_NEAREST)cv.imshow('INTER_NEA...原创 2019-06-04 21:13:57 · 329 阅读 · 0 评论 -
OpenCV-Day-013:图像翻转
代码import cv2 as cvsrc = cv.imread('./test.png', cv.IMREAD_COLOR)cv.imshow('input', src)dst1 = cv.flip(src, 0)cv.imshow('X-flip', dst1)dst2 = cv.flip(src, 1)cv.imshow('Y-flip', dst2)dst3 = cv....原创 2019-06-04 19:00:23 · 227 阅读 · 0 评论 -
OpenCV-Day-012:视频文件的读写
代码import cv2 as cvimport numpy as npcapture = cv.VideoCapture(0)width = 640height = 480capture.set(cv.CAP_PROP_FRAME_HEIGHT, height)capture.set(cv.CAP_PROP_FRAME_WIDTH, width)out = cv.VideoWr...原创 2019-06-04 17:40:37 · 247 阅读 · 0 评论 -
OpenCV-Day-011:像素归一化
预备知识opencv中提供了四种归一化的方法:NORM_MINMAXNORM_INFNORM_L1NORM_L2最常用的就是NORM_MINMAX归一化方法L1 归一化(依据是:和为1)2.0+8.0+10.0=202.0 0.1 (2.0/20.0)8.0 0.4 (8.0/20.0)10.0 ...原创 2019-06-03 23:57:49 · 2293 阅读 · 0 评论 -
OpenCV-Day-010:图像像素值统计
代码import cv2 as cvimport numpy as npsrc = cv.imread('./bird.jpg', cv.IMREAD_GRAYSCALE)cv.imshow('input', src)# 最大最小值和相应的位置min, max, minLoc, maxLoc = cv.minMaxLoc(src)print('min:%.2f,max:%.2f,'...原创 2019-05-25 10:25:56 · 2882 阅读 · 0 评论 -
OpenCV-Day-009:不同色彩空间的转换
前言一、常见色彩空间RGB色彩空间HSV色彩空间HSV是一种比较直观的颜色模型,所以在许多图像编辑工具中应用比较广泛,这个模型中颜色的参数分别是:色调(H, Hue),饱和度(S,Saturation),明度(V, Value)。2.利用openCV中cvSplit函数的在选择图像IPL_DEPTH_32F类型时,H取值范围是0-360,S取值范围是0-1(0%-100%),V取值范围...原创 2019-05-24 20:37:59 · 396 阅读 · 0 评论 -
OpenCV-Day-008:通道的合并与分离
代码实验结果解释原创 2019-05-23 10:50:09 · 210 阅读 · 0 评论 -
OpenCV-Day-007:图像像素的逻辑操作
代码import cv2 as cvimport numpy as npsrc = cv.imread('./0.jpg', cv.IMREAD_COLOR)cv.imshow('raw', src)print(src.shape)src2 = np.zeros(shape=(500, 500, 3), dtype=np.uint8)src2[100:200, 100:200, 1...原创 2019-05-23 09:51:52 · 388 阅读 · 0 评论 -
OpenCV-Day-006: Look Up Table(LUT)查找表
代码import cv2 as cvsrc = cv.imread('./0.jpg', cv.IMREAD_COLOR)cv.namedWindow('raw', cv.WINDOW_AUTOSIZE)cv.imshow('raw', src)dvt = cv.applyColorMap(src, cv.COLORMAP_HOT)cv.imshow('Hot', dvt)dvt2...原创 2019-05-22 11:45:43 · 674 阅读 · 0 评论 -
OpenCV-Day-005:图像像素的算术操作
代码import cv2 as cvimport numpy as npsrc = cv.imread('./0.jpg', cv.IMREAD_COLOR)cv.imshow('raw', src)src2 = cv.imread('./gray.jpg', cv.IMREAD_COLOR)cv.imshow('gray', src2)add_result = np.zeros...原创 2019-05-22 11:20:14 · 241 阅读 · 0 评论 -
OpenCV-Day-004:图像像素的读写操作
代码import cv2 as cvsrc = cv.imread('./0.jpg', cv.IMREAD_COLOR)cv.namedWindow('rawPicture', cv.WINDOW_AUTOSIZE)cv.imshow('rawPicture', src)h, w, ch = src.shapeprint('h,w,ch', h, w, ch)for row i...原创 2019-05-21 23:14:01 · 306 阅读 · 0 评论 -
OpenCV-Day-003:图像对象的创建与赋值
代码import cv2 as cvimport numpy as npsrc = cv.imread('./0.jpg', cv.IMREAD_COLOR)cv.namedWindow('rawPicture', cv.WINDOW_AUTOSIZE)cv.imshow('rawPicture', src)temp = np.copy(src)cv.imshow('temp',...原创 2019-05-21 22:56:04 · 408 阅读 · 0 评论 -
OpenCV-Day-002:图像色彩空间转换
代码import cv2 as cvsrc = cv.imread('0.jpg')cv.namedWindow('RGB', cv.WINDOW_AUTOSIZE)cv.imshow('RGB', src)Gray = cv.cvtColor(src, cv.COLOR_RGB2GRAY)cv.imshow('Gray', Gray)cv.imwrite('gray.jpg',...原创 2019-05-20 17:39:41 · 241 阅读 · 0 评论 -
OpenCV-Day-001:图像读取与显示
代码import cv2 as cvsrc = cv.imread('./0.jpg')cv.namedWindow('newWindow', cv.WINDOW_AUTOSIZE)cv.imshow('newWindow', src)cv.waitKey(1000)cv.destroyAllWindows()解释:一、cv.imread()是读取图片的函数,参数是图片所在的...原创 2019-05-20 11:39:42 · 375 阅读 · 0 评论 -
OpenCV-Day-000:系统化学习之路
从今天开始学习Opencv,因为工作需要,所以提前学习,加油!!!原创 2019-05-20 11:06:22 · 463 阅读 · 2 评论