Commit 01ccffa1 authored by Jay's avatar Jay Committed by jay
Browse files

FLANN based matcher class in place.

parent cdf2445a

autocnet/matcher/matcher.py

+11 −47
Original line number Diff line number Diff line 
import cv2

import numpy as np



import pandas as pd

from autocnet.graph.network import CandidateGraph



FLANN_INDEX_KDTREE = 1
@@ -26,7 +26,7 @@ class FlannMatcher(object): 
    def __init__(self, flann_parameters=DEFAULT_FLANN_PARAMETERS):

        self._flann_matcher = cv2.FlannBasedMatcher(flann_parameters, {})

        self.image_indices = {}

        self.image_index_counter = 0  # OpenCv DMatch.imgIdx are one based

        self.image_index_counter = 0



    def add(self, descriptor, key):

        """
@@ -64,52 +64,16 @@ class FlannMatcher(object): 


        Returns

        -------

         :

        matched : dataframe

                  containing matched points

        """

        print(self.image_indices)

        matches = self._flann_matcher.knnMatch(descriptor, k=k)

        print(dir(self._flann_matcher))

        print(matches)

        matched = []

        for m in matches:

            print(m[0].imgIdx, m[0].queryIdx, m[0].trainIdx)

            #print(self.image_indices[m[0].imgIdx])





def match_features(iterable, flann_parameters=DEFAULT_FLANN_PARAMETERS):

    """

    Iterate over all nodes in the graph, create a single KDTree,

    and then apply a FLANN matcher.



    Parameters

    ----------

    iterable : iterable

            An iterable of descriptors



    flann_parameters : dict

                       of parameters for the FLANN matcher





    Returns

    -------



    """

    flann_matcher = create_flann_matcher(iterable, flann_parameters)





def create_flann_matcher(iterable, flann_parameters):



    flann_matcher = cv2.FlannBasedMatcher(flann_parameters, {})



    try:

        if isinstance(iterable, CandidateGraph):

            training_descriptors = [i[1]['descriptors'] for i in iterable.nodes_iter(data=True)]

        elif isinstance(iterable[0], np.ndarray):

            training_descriptors = iterable

        else:

            raise(TypeError, 'Unsupported iterable passed to match_features')

    except TypeError:

        print('Object is not iterable')

            matched.append((self.image_indices[m[1].imgIdx],

                  m[1].queryIdx,

                  m[1].trainIdx,

                  m[1].distance))

        return pd.DataFrame(matched, columns=['matched_to', 'queryIdx',

                                              'trainIdx', 'distance'])

    flann_matcher.add(training_descriptors)

    flann_matcher.train()

    return flann_matcher
 
 No newline at end of file

autocnet/matcher/tests/test_matcher.py

+7 −5
Original line number Diff line number Diff line
@@ -23,7 +23,7 @@ class TestMatcher(unittest.TestCase): 
        self.fd['AS15-M-0296_SML.png'] = sift.detectAndCompute(im1, None)

        self.fd['AS15-M-0297_SML.png'] = sift.detectAndCompute(im2, None)



    def test_flann_match(self):

    def test_flann_match_two_images(self):



        fmatcher = matcher.FlannMatcher()

        truth_image_indices = {}
@@ -37,11 +37,13 @@ class TestMatcher(unittest.TestCase): 


        fmatcher.train()



        matches = fmatcher.query(self.fd['AS15-M-0296_SML.png'][1], k=2)

        #self.assertEqual(10, len(matches))

        #self.assertEqual(2, len(matches[0]))

        matched = fmatcher.query(self.fd['AS15-M-0296_SML.png'][1], k=2)

        self.assertEqual(10, len(matched))



        # Check that self neighbors are being omitted

        distance = matched['distance']

        self.assertFalse(distance[distance == 0].any())



        self.assertTrue(False)



    def tearDown(self):

        pass
 
 No newline at end of file