python+opencv实现相似图片的搜索

在学习时：http://python.jobbole.com/80860/

在这里对上面给出的链接中的代码进行整理和修改了下，图像搜索的原理，以及搜索的大致步骤和想法，在原博主文章中已经讲解的很详细了，在这里我就不写了，对于上面链接中的代码，有些地方是需要改动的

先贴出我的代码：

直接上代码：

color_descriptor.py

# -*- coding: utf-8 -*-
# !/usr/bin/env python
# @Time    : 2018/11/6 15:17
# @Author  : xhh
# @Desc    : 颜色空间特征提取器
# @File    : color_descriptor.py
# @Software: PyCharm
import cv2
import numpyclass ColorDescriptor:__slot__ = ["bins"]def __init__(self, bins):self.bins = bins# 得到图片的色彩直方图，mask为图像处理区域的掩模def getHistogram(self, image, mask, isCenter):# 利用OpenCV中的calcHist得到图片的直方图imageHistogram = cv2.calcHist([image], [0, 1, 2], mask, self.bins, [0, 180, 0, 256, 0, 256])# 标准化(归一化)直方图normalizeimageHistogram = cv2.normalize(imageHistogram, imageHistogram).flatten()# isCenter判断是否为中间点，对色彩特征向量进行加权处理if isCenter:weight = 5.0  # 权重记为0.5for index in range(len(imageHistogram)):imageHistogram[index] *= weightreturn imageHistogram# 将图像从BGR色彩空间转换为HSV色彩空间def describe(self, image):image = cv2.cvtColor(image, cv2.COLOR_BGR2HSV)features = []# 获取图片的中心点和图片的大小height, width = image.shape[0], image.shape[1]centerX, centerY = int(width * 0.5), int(height * 0.5)# initialize mask dimension# 生成左上、右上、左下、右下、中心部分的掩模。# 中心部分掩模的形状为椭圆形。这样能够有效区分中心部分和边缘部分，从而在getHistogram()方法中对不同部位的色彩特征做加权处理。segments = [(0, centerX, 0, centerY), (0, centerX, centerY, height), (centerX, width, 0, centerY), (centerX, width, centerY, height)]# 初始化中心部分axesX, axesY = int(width * 0.75) / 2, int (height * 0.75) / 2ellipseMask = numpy.zeros([height, width], dtype="uint8")cv2.ellipse(ellipseMask, (int(centerX), int(centerY)), (int(axesX), int(axesY)), 0, 0, 360, 255, -1)#cv2.ellipse(ellipMask, (int(cX), int(cY)), (int(axesX), int(axesY)), 0, 0, 360, 255, -1)# 初始化边缘部分for startX, endX, startY, endY in segments:cornerMask = numpy.zeros([height, width], dtype="uint8")cv2.rectangle(cornerMask, (startX, startY), (endX, endY), 255, -1)cornerMask = cv2.subtract(cornerMask, ellipseMask)# 得到边缘部分的直方图imageHistogram = self.getHistogram(image, cornerMask, False)features.append(imageHistogram)# 得到中心部分的椭圆直方图imageHistogram = self.getHistogram(image, ellipseMask, True)features.append(imageHistogram)# 得到最终的特征值return features

structure_descriptor.py

# -*- coding: utf-8 -*-
# !/usr/bin/env python
# @Time    : 2018/11/6 15:18
# @Author  : xhh
# @Desc    : 构图空间提取器
# @File    : structure_descriptor.py
# @Software: PyCharm
import cv2
# 将图片进行归一化处理，返回HSV色彩空间矩阵
class StructureDescriptor:__slot__ = ["dimension"]def __init__(self, dimension):self.dimension = dimensiondef describe(self, image):image = cv2.resize(image, self.dimension, interpolation=cv2.INTER_CUBIC)# 将图片转化为BGR图片转化为HSV格式image = cv2.cvtColor(image, cv2.COLOR_BGR2HSV)# print(image)return image

searchEngine.py

# -*- coding: utf-8 -*-
# !/usr/bin/env python
# @Time    : 2018/11/6 15:21
# @Author  : xhh
# @Desc    :
# @File    : searchEngine.py
# @Software: PyCharm
import color_descriptor
import structure_descriptor
import searcher
import argparse
import cv2# 构造解析函数
searchArgParser = argparse.ArgumentParser()
searchArgParser.add_argument("-c", "--colorindex", required = True, help = "Path to where the computed color index will be stored")
searchArgParser.add_argument("-s", "--structureindex", required = True, help = "Path to where the computed structure index will be stored")
searchArgParser.add_argument("-q", "--query", required=True, help = "Path to the query image")
searchArgParser.add_argument("-r", "--resultpath", required = True, help = "Path to the result path")
searchArguments = vars(searchArgParser.parse_args())idealBins = (8, 12, 3)
idealDimension = (16, 16)# 传入色彩空间的bins
colorDescriptor = color_descriptor.ColorDescriptor(idealBins)
# 传入构图空间的bins
structureDescriptor = structure_descriptor.StructureDescriptor(idealDimension)queryImage = cv2.imread(searchArguments["query"])
colorIndexPath = searchArguments["colorindex"]
structureIndexPath = searchArguments["structureindex"]
resultPath = searchArguments["resultpath"]queryFeatures = colorDescriptor.describe(queryImage)
queryStructures = structureDescriptor.describe(queryImage)
imageSearcher = searcher.Searcher(colorIndexPath, structureIndexPath)
searchResults = imageSearcher.search(queryFeatures, queryStructures)# 对搜索到的图片进行展示
for imageName, score in searchResults:queryResult = cv2.imread(resultPath + "/" + imageName)cv2.imshow("Result Score: " + str(int(score)) + " (lower is better)", queryResult)cv2.waitKey(0)
cv2.imshow("Query", queryImage)
cv2.waitKey(0)

index.py

# -*- coding: utf-8 -*-
# !/usr/bin/env python
# @Time    : 2018/11/6 15:20
# @Author  : xhh
# @Desc    :
# @File    : index.py
# @Software: PyCharm
import color_descriptor
import structure_descriptor
import glob
import argparse
import cv2# 创建解析函数
searchArgParser = argparse.ArgumentParser()
searchArgParser.add_argument("-d", "--dataset", required = True, help = "Path to the directory that contains the images to be indexed")
searchArgParser.add_argument("-c", "--colorindex", required = True, help = "Path to where the computed color index will be stored")
searchArgParser.add_argument("-s", "--structureindex", required = True, help = "Path to where the computed structure index will be stored")
arguments = vars(searchArgParser.parse_args())idealBins = (8, 12, 3)
colorDesriptor = color_descriptor.ColorDescriptor(idealBins)
output = open(arguments["colorindex"], "w")# 色彩空间的特征存储
for imagePath in glob.glob(arguments["dataset"] + "/*.png"):imageName = imagePath[imagePath.rfind("\\") + 1 : ]   # 这里也是需要修改的image = cv2.imread(imagePath)features = colorDesriptor.describe(image)# 将色彩空间的特征写入到csv文件中去features = [str(feature).replace("\n", "") for feature in features]output.write("%s,%s\n" % (imageName, ",".join(features)))
# close index file
output.close()idealDimension = (16, 16)
structureDescriptor = structure_descriptor.StructureDescriptor(idealDimension)
output = open(arguments["structureindex"], "w")# 构图空间的色彩特征存储
for imagePath in glob.glob("dataset" + "/*.png"):# imageName = imagePath[imagePath.rfind("/") + 1 : ]   # 这里是需要修改的imageName = imagePath[imagePath.rfind("\\") + 1 : ]image = cv2.imread(imagePath)structures = structureDescriptor.describe(image)# 将构图空间的色彩特征写入到文件中去  write structures to filestructures = [str(structure).replace("\n", "") for structure in structures]output.write("%s,%s\n" % (imageName, ",".join(structures)))
# close index file
output.close()

searcher.py

# -*- coding: utf-8 -*-
# !/usr/bin/env python
# @Time    : 2018/11/6 15:19
# @Author  : xhh
# @Desc    : 图片搜索内核
# @File    : searcher.py
# @Software: PyCharm
import numpy
import csv
import reclass Searcher:# colorIndexPath色彩空间特征索引表路径，structureIndexPath结构特征索引表路径__slot__ = ["colorIndexPath", "structureIndexPath"]def __init__(self, colorIndexPath, structureIndexPath):self.colorIndexPath, self.structureIndexPath = colorIndexPath, structureIndexPath# 计算色彩空间的距离，卡方相似度计算def solveColorDistance(self, features, queryFeatures, eps = 1e-5):distance = 0.5 * numpy.sum([((a - b) ** 2) / (a + b + eps) for a, b in zip(features, queryFeatures)])return distance# 计算构图空间的距离def solveStructureDistance(self, structures, queryStructures, eps = 1e-5):distance = 0normalizeRatio = 5e3for index in range(len(queryStructures)):for subIndex in range(len(queryStructures[index])):a = structures[index][subIndex]b = queryStructures[index][subIndex]distance += (a - b) ** 2 / (a + b + eps)return distance / normalizeRatiodef searchByColor(self, queryFeatures):searchResults = {}with open(self.colorIndexPath) as indexFile:reader = csv.reader(indexFile)for line in reader:features = []for feature in line[1:]:feature = feature.replace("[", "").replace("]", "")findStartPosition = 0feature = re.split("\s+", feature)rmlist = []for index, strValue in enumerate(feature):if strValue == "":rmlist.append(index)for _ in range(len(rmlist)):currentIndex = rmlist[-1]rmlist.pop()del feature[currentIndex]feature = [float(eachValue) for eachValue in feature]features.append(feature)distance = self.solveColorDistance(features, queryFeatures)searchResults[line[0]] = distanceindexFile.close()# print "feature", sorted(searchResults.iteritems(), key = lambda item: item[1], reverse = False)return searchResultsdef transformRawQuery(self, rawQueryStructures):queryStructures = []for substructure in rawQueryStructures:structure = []for line in substructure:for tripleColor in line:structure.append(float(tripleColor))queryStructures.append(structure)return queryStructuresdef searchByStructure(self, rawQueryStructures):searchResults = {}queryStructures = self.transformRawQuery(rawQueryStructures)with open(self.structureIndexPath) as indexFile:reader = csv.reader(indexFile)for line in reader:structures = []for structure in line[1:]:structure = structure.replace("[", "").replace("]", "")structure = re.split("\s+", structure)if structure[0] == "":structure = structure[1:]structure = [float(eachValue) for eachValue in structure]structures.append(structure)distance = self.solveStructureDistance(structures, queryStructures)searchResults[line[0]] = distanceindexFile.close()# print "structure", sorted(searchResults.iteritems(), key = lambda item: item[1], reverse = False)return searchResultsdef search(self, queryFeatures, rawQueryStructures, limit = 6):featureResults = self.searchByColor(queryFeatures)structureResults = self.searchByStructure(rawQueryStructures)results = {}for key, value in featureResults.iteritems():results[key] = value + structureResults[key]results = sorted(results.iteritems(), key = lambda item: item[1], reverse = False)return results[ : limit]

在这里返回的结果集只有一个值：

运行：我是在windows10下项目下直接cmd的

步骤：

1.提取特征图片的直方图特征，以及图片的构图空间特征，index.csv

python index.py --dataset dataset --colorindex color——index.csv --structure structure_index.csv

2.图片查找：

python searchEngine.py -c color_index.csv -s structure_index.csv -r dataset -q query/mm.png

查找出来的图片，感觉差不多

注意：

对于灰度图，在这里效果很差，，我对一个图片进行处理转化为灰度图处理了之后，在进行相似图片的查找直接放图吧

查找出来的图片：

效果很差

以上的相似图片的搜索，利用的直方图的差异来查找，，直方图只是对色彩空间的敏感度高，后续还需要要改进，也利用了hash值来计算，后续在贴出来吧。

大家可以关注我和我小伙伴的公众号~~~这里有我和我的小伙伴不定时的更新一些python技术资料哦！！大家也可以留言，讨论一下技术问题，希望大家多多支持，关注一下啦，谢谢大家啦~~

python+opencv实现相似图片的搜索相关推荐

python+opencv将表格图片按照表格框线分割、识别
如下小程序为使用python+opencv将表格图片,按照表格进行分割,并识别分割后的子图片中的文字,希望对需要的小伙伴有一些些帮助.具体的实现见如下代码. # -*- coding: utf-8 - ...
python opencv 如何给图片添加文字？cv2.putText() PIL
参考文章1:python如何在图片上添加文字(中文和英文)Python在图片上添加文字的两种方法:OpenCV和PIL 参考文章2:python之------如何在图片上面添加文字(多种类型的文字)[ ...
python+opencv人脸识别-图片处理+代码
首先安装opencv: cmd 然后输入:install opencv-python 刚接触人脸识别,正在学习:代码是从绘制图片到人脸识别,里面的jpg图片找个有一个人脸图片就行,放在项目的根目录下就 ...
python opencv轻松去图片水印
背景最近有一个需要为图片去水印的需求,于是各种折腾开始. 背景了解图片标准图片使用RGB编码,RGB色彩模式是工业界的一种颜色标准,是通过对红(R).绿(G).蓝(B)三个颜色通道的变化以及它们相 ...
Python+OpenCV批量给图片加噪声
利用python和opencv批量给图像加噪声(椒盐噪声.高斯噪声.随机噪声) 导入头文件 import os import cv2 import numpy as np import random ...
python opencv numpy旋转图片
python旋转图片背景在图像处理中,有的时候会有对图片进行角度旋转的处理,尤其是在计算机视觉中对于图像扩充,旋转角度扩充图片是一种常见的处理.这种旋转图片的应用场景也比较多,比如用户上传图片是竖 ...
Python+Opencv实现把图片转为视频
1. 安装Opencv包在Python命令行输入如下命令(如果你使用的Anaconda,直接进入Anaconda Prompt键入命令即可.如果你不知道Anaconda是什么,可以参考王树义老师的文 ...
python opencv 三维重建_【python+opencv实现基于图片序列的三维重建】 - #1
2015年09月05 - 三维重建一直是机器视觉研究的热门方向,比如,基于双目视觉,单目视觉,多视几何,光场三维重建等等. 每一种方法都有其有点和局限性.单目视觉需要拍摄多幅图像,并且在拍摄过程中需要 ...
Python+opencv裁剪/截取图片的几种方式
文章目录前言一.手动单张裁剪/截取二.根据图片的位置坐标进行裁剪/截取三.根据xml文件截取/裁剪目标四.opencv获取边缘并根据bounding box截取/裁剪目标五.用YOLO目标 ...
Python OpenCV 实现对图片旋转矩形的裁剪
前言: 最近需要实现一个在一张大图上给定一个带方向的点,以这个点为中心,以方向为角度裁剪出一个矩形区域.在网上查到很多方法都是先对大图像进行旋转,再从中切片取出目标区域.不失为一个好办法,但是我要处理 ...

python+opencv实现相似图片的搜索

python+opencv实现相似图片的搜索相关推荐

最新文章

热门文章