Documentation

Create a new cascade classifier. Returns Nothing if the classifier is empty after initialization. This usually means that the file could not be loaded (e.g. it doesn't exist, is corrupt, etc.)

Example:

cascadeClassifierArnold
  :: forall (width    :: Nat)
            (height   :: Nat)
            (channels :: Nat)
            (depth    :: *  )
   . (Mat (ShapeT [height, width]) ('S channels) ('S depth) ~ Arnold_small)
  => IO (Mat (ShapeT [height, width]) ('S channels) ('S depth))
cascadeClassifierArnold = do
    -- Create two classifiers from data files.
    Just ccFrontal <- newCascadeClassifier "data/haarcascade_frontalface_default.xml"
    Just ccEyes    <- newCascadeClassifier "data/haarcascade_eye.xml"
    -- Detect some features.
    let eyes  = ccDetectMultiscale ccEyes    arnoldGray
        faces = ccDetectMultiscale ccFrontal arnoldGray
    -- Draw the result.
    pure $ exceptError $
      withMatM (Proxy :: Proxy [height, width])
               (Proxy :: Proxy channels)
               (Proxy :: Proxy depth)
               white $ \imgM -> do
        void $ matCopyToM imgM (V2 0 0) arnold_small Nothing
        forM_ eyes  $ \eyeRect  -> lift $ rectangle imgM eyeRect  blue  2 LineType_8 0
        forM_ faces $ \faceRect -> lift $ rectangle imgM faceRect green 2 LineType_8 0
  where
    arnoldGray = exceptError $ cvtColor bgr gray arnold_small

    ccDetectMultiscale cc = cascadeClassifierDetectMultiScale cc Nothing Nothing minSize maxSize

    minSize = Nothing :: Maybe (V2 Int32)
    maxSize = Nothing :: Maybe (V2 Int32)

Special version which returns bounding rectangle, rejectLevels, and levelWeights