Python构建图像分类识别器的方法


Posted in Python onJanuary 12, 2019

机器学习用在图像识别是非常有趣的话题。

我们可以利用OpenCV强大的功能结合机器学习算法实现图像识别系统。

首先,输入若干图像,加入分类标记。利用向量量化方法将特征点进行聚类,并得出中心点,这些中心点就是视觉码本的元素。

其次,利用图像分类器将图像分到已知的类别中,ERF(极端随机森林)算法非常流行,因为ERF具有较快的速度和比较精确的准确度。我们利用决策树进行正确决策。

最后,利用训练好的ERF模型后,创建目标识别器,可以识别未知图像的内容。

当然,这只是雏形,存在很多问题:

界面不友好。

准确率如何保证,如何调整超参数,只有认真研究算法机理,才能真正清除内部实现机制后给予改进。

下面,上代码:

import os

import sys
import argparse
import json
import cv2
import numpy as np
from sklearn.cluster import KMeans
# from star_detector import StarFeatureDetector
from sklearn.ensemble import ExtraTreesClassifier
from sklearn import preprocessing

try:
 import cPickle as pickle #python 2
except ImportError as e:
 import pickle #python 3

def load_training_data(input_folder):
 training_data = []
 if not os.path.isdir(input_folder):
  raise IOError("The folder " + input_folder + " doesn't exist")
  
 for root, dirs, files in os.walk(input_folder):
  for filename in (x for x in files if x.endswith('.jpg')):
   filepath = os.path.join(root, filename)
   print(filepath)
   object_class = filepath.split('\\')[-2]
   print("object_class",object_class)
   training_data.append({'object_class': object_class, 'image_path': filepath})
     
 return training_data
class StarFeatureDetector(object):
 def __init__(self):
  self.detector = cv2.xfeatures2d.StarDetector_create()
 def detect(self, img):
  return self.detector.detect(img)

class FeatureBuilder(object):
 def extract_features(self, img):
  keypoints = StarFeatureDetector().detect(img)
  keypoints, feature_vectors = compute_sift_features(img, keypoints)
  return feature_vectors
 def get_codewords(self, input_map, scaling_size, max_samples=12):
  keypoints_all = []
  count = 0
  cur_class = ''
  for item in input_map:
   if count >= max_samples:
    if cur_class != item['object_class']:
     count = 0
    else:
     continue
   count += 1
   if count == max_samples:
    print ("Built centroids for", item['object_class'])

   cur_class = item['object_class']
   img = cv2.imread(item['image_path'])
   img = resize_image(img, scaling_size)
   num_dims = 128
   feature_vectors = self.extract_features(img)
   keypoints_all.extend(feature_vectors)

  kmeans, centroids = BagOfWords().cluster(keypoints_all)
  return kmeans, centroids
class BagOfWords(object):
 def __init__(self, num_clusters=32):
  self.num_dims = 128
  self.num_clusters = num_clusters
  self.num_retries = 10

 def cluster(self, datapoints):
  kmeans = KMeans(self.num_clusters, 
      n_init=max(self.num_retries, 1),
      max_iter=10, tol=1.0)
  res = kmeans.fit(datapoints)
  centroids = res.cluster_centers_
  return kmeans, centroids

 def normalize(self, input_data):
  sum_input = np.sum(input_data)

  if sum_input > 0:
   return input_data / sum_input
  else:
   return input_data
 def construct_feature(self, img, kmeans, centroids):
  keypoints = StarFeatureDetector().detect(img)
  keypoints, feature_vectors = compute_sift_features(img, keypoints)
  labels = kmeans.predict(feature_vectors)
  feature_vector = np.zeros(self.num_clusters)

  for i, item in enumerate(feature_vectors):
   feature_vector[labels[i]] += 1

  feature_vector_img = np.reshape(feature_vector, ((1, feature_vector.shape[0])))
  return self.normalize(feature_vector_img)
# Extract features from the input images and 
# map them to the corresponding object classes
def get_feature_map(input_map, kmeans, centroids, scaling_size):
 feature_map = []
 for item in input_map:
  temp_dict = {}
  temp_dict['object_class'] = item['object_class']
 
  print("Extracting features for", item['image_path'])
  img = cv2.imread(item['image_path'])
  img = resize_image(img, scaling_size)

  temp_dict['feature_vector'] = BagOfWords().construct_feature(img, kmeans, centroids)
  if temp_dict['feature_vector'] is not None:
   feature_map.append(temp_dict)
 return feature_map

# Extract SIFT features
def compute_sift_features(img, keypoints):
 if img is None:
  raise TypeError('Invalid input image')

 img_gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
 keypoints, descriptors = cv2.xfeatures2d.SIFT_create().compute(img_gray, keypoints)
 return keypoints, descriptors

# Resize the shorter dimension to 'new_size' 
# while maintaining the aspect ratio
def resize_image(input_img, new_size):
 h, w = input_img.shape[:2]
 scaling_factor = new_size / float(h)

 if w < h:
  scaling_factor = new_size / float(w)

 new_shape = (int(w * scaling_factor), int(h * scaling_factor))
 return cv2.resize(input_img, new_shape)

def build_features_main():
 data_folder = 'training_images\\'
 scaling_size = 200
 codebook_file='codebook.pkl'
 feature_map_file='feature_map.pkl'
 # Load the training data
 training_data = load_training_data(data_folder)

 # Build the visual codebook
 print("====== Building visual codebook ======")
 kmeans, centroids = FeatureBuilder().get_codewords(training_data, scaling_size)
 if codebook_file:
  with open(codebook_file, 'wb') as f:
   pickle.dump((kmeans, centroids), f)
 
 # Extract features from input images
 print("\n====== Building the feature map ======")
 feature_map = get_feature_map(training_data, kmeans, centroids, scaling_size)
 if feature_map_file:
  with open(feature_map_file, 'wb') as f:
   pickle.dump(feature_map, f)
# --feature-map-file feature_map.pkl --model- file erf.pkl
#----------------------------------------------------------------------------------------------------------
class ERFTrainer(object):
 def __init__(self, X, label_words):
  self.le = preprocessing.LabelEncoder()
  self.clf = ExtraTreesClassifier(n_estimators=100,
    max_depth=16, random_state=0)

  y = self.encode_labels(label_words)
  self.clf.fit(np.asarray(X), y)

 def encode_labels(self, label_words):
  self.le.fit(label_words)
  return np.array(self.le.transform(label_words), dtype=np.float32)

 def classify(self, X):
  label_nums = self.clf.predict(np.asarray(X))
  label_words = self.le.inverse_transform([int(x) for x in label_nums])
  return label_words
#------------------------------------------------------------------------------------------

class ImageTagExtractor(object):
 def __init__(self, model_file, codebook_file):
  with open(model_file, 'rb') as f:
   self.erf = pickle.load(f)

  with open(codebook_file, 'rb') as f:
   self.kmeans, self.centroids = pickle.load(f)

 def predict(self, img, scaling_size):
  img = resize_image(img, scaling_size)
  feature_vector = BagOfWords().construct_feature(
    img, self.kmeans, self.centroids)
  image_tag = self.erf.classify(feature_vector)[0]
  return image_tag

def train_Recognizer_main():
 feature_map_file = 'feature_map.pkl'
 model_file = 'erf.pkl'
 # Load the feature map
 with open(feature_map_file, 'rb') as f:
  feature_map = pickle.load(f)
 # Extract feature vectors and the labels
 label_words = [x['object_class'] for x in feature_map]
 dim_size = feature_map[0]['feature_vector'].shape[1]
 X = [np.reshape(x['feature_vector'], (dim_size,)) for x in feature_map]

 # Train the Extremely Random Forests classifier
 erf = ERFTrainer(X, label_words)
 if model_file:
  with open(model_file, 'wb') as f:
   pickle.dump(erf, f)
 #--------------------------------------------------------------------
 # args = build_arg_parser().parse_args()
 model_file = 'erf.pkl'
 codebook_file ='codebook.pkl'
 import os
 rootdir=r"F:\airplanes"
 list=os.listdir(rootdir)
 for i in range(0,len(list)):
  path=os.path.join(rootdir,list[i])
  if os.path.isfile(path):
   try:
    print(path)
    input_image = cv2.imread(path)
    scaling_size = 200
    print("\nOutput:", ImageTagExtractor(model_file,codebook_file)\
      .predict(input_image, scaling_size))
   except:
    continue
 #-----------------------------------------------------------------------
build_features_main()
train_Recognizer_main()

以上这篇Python构建图像分类识别器的方法就是小编分享给大家的全部内容了,希望能给大家一个参考,也希望大家多多支持三水点靠木。

Python 相关文章推荐
Python smallseg分词用法实例分析
May 28 Python
Python ldap实现登录实例代码
Sep 30 Python
Python使用正则表达式抓取网页图片的方法示例
Apr 21 Python
用python结合jieba和wordcloud实现词云效果
Sep 05 Python
python网络爬虫之如何伪装逃过反爬虫程序的方法
Nov 23 Python
使用pygame模块编写贪吃蛇的实例讲解
Feb 05 Python
python 循环读取txt文档 并转换成csv的方法
Oct 26 Python
Python定时任务工具之APScheduler使用方式
Jul 24 Python
python IDLE添加行号显示教程
Apr 25 Python
python实现秒杀商品的微信自动提醒功能(代码详解)
Apr 27 Python
django queryset 去重 .distinct()说明
May 19 Python
如何表示python中的相对路径
Jul 08 Python
使用python opencv对目录下图片进行去重的方法
Jan 12 #Python
python变量赋值方法(可变与不可变)
Jan 12 #Python
python交换两个变量的值方法
Jan 12 #Python
对Pyhon实现静态变量全局变量的方法详解
Jan 11 #Python
浅谈Python中的全局锁(GIL)问题
Jan 11 #Python
Python 实现子类获取父类的类成员方法
Jan 11 #Python
python使用xlrd模块读取xlsx文件中的ip方法
Jan 11 #Python
You might like
php中记录用户访问过的产品,在cookie记录产品id,id取得产品信息
2011/05/04 PHP
PHP中strtotime函数使用方法详解
2011/11/27 PHP
PHP 代码规范小结
2012/03/08 PHP
thinkPHP5.0框架模块设计详解
2017/03/18 PHP
如何让PHP编码更加好看利于阅读
2019/05/12 PHP
laravel框架中路由设置,路由参数和路由命名实例分析
2019/11/23 PHP
JavaScript 面向对象的 私有成员和公开成员
2010/05/13 Javascript
event对象获取方法总结在google浏览器下测试
2013/11/03 Javascript
javascript loadScript异步加载脚本示例讲解
2013/11/14 Javascript
avascript中的自执行匿名函数应用示例
2014/09/15 Javascript
jQuery拖拽插件gridster使用指南
2015/04/21 Javascript
JavaScript 弹出子窗体并返回结果到父窗体的实现代码
2016/05/28 Javascript
基于JavaScript实现树形下拉框
2016/08/10 Javascript
Angular+Node生成随机数的方法
2017/06/16 Javascript
浅析vue-router jquery和params传参(接收参数)$router $route的区别
2018/08/03 jQuery
解决jquery validate 验证不通过后验证正确的信息仍残留在label上的方法
2019/08/27 jQuery
原生js+css实现tab切换功能
2020/09/17 Javascript
PyCharm在win10的64位系统安装实例
2017/11/26 Python
详解重置Django migration的常见方式
2019/02/15 Python
pandas-resample按时间聚合实例
2019/12/27 Python
浅析python函数式编程
2020/09/26 Python
乐天旅游香港网站:日本饭店预订
2017/11/29 全球购物
澳大利亚厨房和家用电器购物网站:Bing Lee
2021/01/11 全球购物
应用服务器有那些
2012/01/19 面试题
数控专业毕业生求职信范文
2013/09/21 职场文书
工程力学专业毕业生求职信
2013/10/06 职场文书
室内设计专业个人的自我评价
2013/10/19 职场文书
简历自荐信
2013/12/02 职场文书
建筑系毕业生自我鉴定
2014/01/24 职场文书
2014年寒假社会实践活动心得体会
2014/04/07 职场文书
职业道德模范事迹材料
2014/08/24 职场文书
党员个人公开承诺书
2014/08/29 职场文书
环境卫生工作汇报材料
2014/10/28 职场文书
金秋助学感谢信
2015/01/21 职场文书
亮剑观后感500字
2015/06/05 职场文书
2016年“5.12”国际护士节活动总结
2016/04/06 职场文书