Mac/Windows differences using ORB feature detector.

edit

Per the bug reporting instructions I'm asking about this first before submitting a bug report. I'm new to ORB so I wanted to make sure I'm not doing anything incorrect first.

The following code produces difference results when run on Mac and Windows.

Windows 10 Pro 64

Mac OSX 10.11.1

OpenCV 3.0.0

Built with MS Visual Studio 2013. x64.

Built with clang 7.0 x64

    cv::Mat a, b; // 8 bit grayscale images loaded elsewhere. CV_8UC1   

    cv::Ptr<cv::ORB> detector = cv::ORB::create(900, 1.2, 8, 15);

std::vector<cv::KeyPoint> keypoints_object, keypoints_scene;

detector->detect(a, keypoints_object);
detector->detect(b, keypoints_scene);
//#ifdef DEBUG
printf("!!!!!!!!!!!%d\n", keypoints_object.size());
printf("!!!!!!!!!!!%d\n", keypoints_scene.size());
//#endif
//-- Step 2: Calculate descriptors (feature vectors)
cv::Ptr<cv::BRISK> extractor = cv::BRISK::create(30, 3, 1.0f);

cv::Mat descriptors_object, descriptors_scene;

extractor->compute(a, keypoints_object, descriptors_object);
extractor->compute(b, keypoints_scene, descriptors_scene);

descriptors_object.convertTo(descriptors_object, CV_32F);
descriptors_scene.convertTo(descriptors_scene, CV_32F);

//-- Step 3: Matching descriptor vectors using FLANN matcher

cv::FlannBasedMatcher matcher;
std::vector<cv::DMatch> matches;
matcher.match(descriptors_object, descriptors_scene, matches);
double max_dist = 0;
double min_dist = 300;

//-- Quick calculation of max and min distances between keypoints
for (int i = 0; i < descriptors_object.rows; i++) {
    double dist = matches[i].distance;
    if (dist < min_dist)
        min_dist = dist;
    if (dist > max_dist)
        max_dist = dist;
}
//#ifdef DEBUG
printf("-- Max dist : %f \n", max_dist);
printf("-- Min dist : %f \n", min_dist);

On Windows my output is

-- Max dist : 876.513550

-- Min dist : 21.587032

On Mac my output is

-- Max dist : 906.911426

-- Min dist : 21.587032

Same inputs.

Thanks for any insight.