feature matching update

CMakeLists.txt

@@ -24,6 +24,6 @@ include_directories(
 
 ## Add folders to be run by python nosetests
 # catkin_add_nosetests(test)
-catkin_install_python(PROGRAMS scripts/Camera.py scripts/featureExtractor.py scripts/vs_nodeHandler.py scripts/Controller.py
- DESTINATION ${CATKIN_PACKAGE_BIN_DESTINATION}
-)
+# catkin_install_python(PROGRAMS scripts/vs_nodeHandler.py
+#  DESTINATION ${CATKIN_PACKAGE_BIN_DESTINATION}
+# )

scripts/featureMatching.py

@@ -1,12 +1,9 @@
 
 import cv2 as cv
-from matplotlib.image import BboxImage
 import numpy as np
 import itertools
-import copy
-from cv_bridge import CvBridge
 import matplotlib.pyplot as plt
-from scipy.signal import find_peaks
+from movingVariance import *
 
 from shapely.geometry import Point
 from shapely.geometry.polygon import Polygon
@@ -45,8 +42,6 @@ def detectNewCropLane(self, mode, rgbImg, greenIDx, binaryMask, wLocs, numofCrop
  _rgbImg = rgbImg.copy()
  # get masked rgb Image (ExG masked)
  rgbMasked = self.maskRgb(rgbImg, binaryMask)
- # find new crop rows and compare to old ones!
-
  # extract keypoints and descriptors from new detected line
  srcKeypoints, srcDescriptors = self.detectTrackingFeatures(rgbMasked)
  # filter in desired window
@@ -56,20 +51,22 @@ def detectNewCropLane(self, mode, rgbImg, greenIDx, binaryMask, wLocs, numofCrop
  # find matches between ref and src keypoints
  matches = self.matchTrackingFeatures(srcKeypoints, srcDescriptors)
  # if not that indicates that the window is between two lines
- print(self.count, len(matches))
-
  # draw keypoints on rgb Image
  self.drawKeyPoints(_rgbImg, srcKeypoints, [0,255,0])
-
+ # collect matches
  self.simDistVec = list(self.simDistVec)
  self.simDistVec.append(len(matches))
-
+ # create recognition signal
  peaks = [] 
- if self.count > 110: 
+ if self.count > 10: 
  # self.simDistVec = np.where(self.simDistVec > 20, np.max(self.simDistVec), self.simDistVec)
- troughs, _ = find_peaks(self.simDistVec, prominence=2, height=5)
- self.simDistVec = np.negative(self.simDistVec)
- peaks, _ = find_peaks(self.simDistVec, prominence=2, height=-5)
+ # troughs, _ = find_peaks(self.simDistVec, prominence=2, height=5)
+ # self.simDistVec = np.negative(self.simDistVec)
+ # peaks, _ = find_peaks(self.simDistVec, prominence=2, height=-5)
+ # compute moving standard deviation
+ mvSignal = movingStd(self.simDistVec)
+ # find positive an negative peaks of the signal
+ peaks, troughs = findPicksTroughths(mvSignal, 0.5)
  plt.plot(self.simDistVec)
  pickPoses = [self.simDistVec[p] for p in peaks]
  troughPoses = [self.simDistVec[p] for p in troughs]
@@ -79,7 +76,6 @@ def detectNewCropLane(self, mode, rgbImg, greenIDx, binaryMask, wLocs, numofCrop
  plt.show() 
 
  self.count+=1
-
  return (len(peaks) >= numofCropRows and len(troughs) >= numofCropRows+1)
 
  def maskRgb(self, rgbImg, mask):

scripts/imageProc.py

@@ -146,7 +146,6 @@ def findCropRows2D(self, rgbImg):
  print(
  '--- Initialisation failed - No lines detected by sweeping window ---')
  self.cropLaneFound = False
-
  return lines, linesROIs
 
  def findCropRowsInMVSignal(self, peaksPos, peaksNeg, mvSignal, lines, linesROIs):