Update faq.markdown

diff --git a/doc/faq.markdown b/doc/faq.markdown
index 3bbaa82..3bf574a 100644 (file)
--- a/doc/faq.markdown
+++ b/doc/faq.markdown
@@ -1,11 +1,23 @@
 Frequently Asked Questions {#faq}
 ==========================
 
--   Q: Example question?
+-   **What is InputArray?**
 
-    A: Example answer
+        It can be seen that almost all OpenCV functions receive InputArray type.
+        What is it, and how can I understand the actual input types of parameters?
 
 
--   Q: Example question?
+    This is the proxy class for passing read-only input arrays into OpenCV functions.
 
-    A: Example answer
+    Inside a function you should use cv::_InputArray::getMat() method to construct
+a matrix header for the array (without copying data). cv::_InputArray::kind() can be used to distinguish Mat from vector<> etc.
+but normally it is not needed.
+
+   for more information see cv::_InputArray
+
+-   **Which is more efficient, use contourArea() or count number of ROI non-zero pixels?**
+
+        In a case where you only want relative areas, which one is faster to compute:
+        calculate a contour area or count the number of ROI non-zero pixels?
+
+    cv::contourArea() uses Green formula (http://en.wikipedia.org/wiki/Green's_theorem) to compute the area, therefore its complexity is O(contour_number_of_vertices). Counting non-zero pixels in the ROI is O(roi_width*roi_height) algorithm, i.e. much slower. Note, however, that because of finite, and quite low, resolution of the raster grid, the two algorithms will give noticeably different results. For large and square-like contours the error will be minimal. For small and/or oblong contours the error can be quite large.