OpenCV3.x Python语言实现---基于文身取证的应用程序示例

0.训练图片数据

1.生成图像描述符generate_descriptors.py

#coding=gbk
import os
import cv2
import numpy as np
from os import walk
from os.path import join
import sys

def create_descriptors(folder):
  files = []
  for (dirpath, dirnames, filenames) in walk(folder):
    files.extend(filenames)
  for f in files:
    save_descriptor(folder, f, cv2.xfeatures2d.SIFT_create())

def save_descriptor(folder, image_path, feature_detector):
  print( "读取中... %s" % image_path)
  if image_path.endswith("npy"):
    return
  img = cv2.imread(join(folder, image_path), 0)
  keypoints, descriptors = feature_detector.detectAndCompute(img, None)
  descriptor_file = image_path.replace("jpg", "npy")
  np.save(join(folder, descriptor_file), descriptors)

dir = sys.argv[1]

create_descriptors(dir)

以命令行方式输入：

2.扫描匹配---检测test.jpg图片（与训练图片放在同一文件夹内）scan_for_matches.py

#coding=gbk
from os.path import join
from os import walk
import numpy as np
import cv2
from sys import argv
from matplotlib import pyplot as plt

# create an array of filenames
folder = argv[1]
query = cv2.imread(join(folder, "test.jpg"), 0)

# create files, images, descriptors globals
files = []
images = []
descriptors = []
for (dirpath, dirnames, filenames) in walk(folder):
  files.extend(filenames)
  for f in files:
    if f.endswith("npy") and f != "test.npy":
      descriptors.append(f)
  print (descriptors)

# create the sift detector
sift = cv2.xfeatures2d.SIFT_create()
query_kp, query_ds = sift.detectAndCompute(query, None)

# create FLANN matcher
FLANN_INDEX_KDTREE = 0
index_params = dict(algorithm = FLANN_INDEX_KDTREE, trees = 5)
search_params = dict(checks = 50)
flann = cv2.FlannBasedMatcher(index_params, search_params)

# minimum number of matches
MIN_MATCH_COUNT = 10

potential_culprits = {}
print (">> 初始化扫描图片...")
for d in descriptors:
  print( "--------- 分析 %s 匹配 ------------" % d)
  matches = flann.knnMatch(query_ds, np.load(join(folder, d)), k =2)
  good = []
  for m,n in matches:
      if m.distance < 0.7*n.distance:
          good.append(m)
  if len(good) > MIN_MATCH_COUNT:
    print ("%s 可匹配! (%d)" % (d, len(good)))
  else:
    print ("%s 不匹配" % d)
  potential_culprits[d] = len(good)

max_matches = None
potential_suspect = None
for culprit, matches in potential_culprits.items():
  if max_matches == None or matches > max_matches:
 max_matches = matches
 potential_suspect = culprit
print ("可能匹配的是 %s" % potential_suspect.replace("npy", "").upper())

#个人增加的代码
culprit_image = potential_suspect.replace("npy", "jpg")
target = cv2.imread(join(folder, culprit_image), 0)
#cv2.imshow('1',target)
#cv2.waitKey()
object_kp, object_ds = sift.detectAndCompute(target, None)
matches_o = flann.knnMatch(query_ds, object_ds, k =2)
well= []
for k,l in matches_o:
if k.distance<0.7*l.distance:
well.append(k)
if len(well)>MIN_MATCH_COUNT:
src_pts = np.float32([ query_kp[k.queryIdx].pt for k in well ]).reshape(-1,1,2)
dst_pts = np.float32([ object_kp[k.trainIdx].pt for k in well ]).reshape(-1,1,2)
M, mask = cv2.findHomography(src_pts, dst_pts, cv2.RANSAC,5.0)
matchesMask = mask.ravel().tolist()
h,w = query.shape
pts = np.float32([ [0,0],[0,h-1],[w-1,h-1],[w-1,0] ]).reshape(-1,1,2)
dst = cv2.perspectiveTransform(pts,M)
img2 = cv2.polylines(target,[np.int32(dst)],True,255,3, cv2.LINE_AA) 
else:
    print( "匹配数量不足 - %d/%d" % (len(well),MIN_MATCH_COUNT))

    matchesMask = None

drawParams=dict(matchColor=(0,255,0),
singlePointColor=None,
matchesMask=matchesMask,
flags=2)

resultImage=cv2.drawMatches(query,query_kp,target,object_kp,well,
None,**drawParams)

plt.imshow(resultImage)

plt.show()

同样以命令行方式输入：

3.结果