Add in LumaA 8 bit and 16 bit images (#74)

LumaA or images that would turn into LumaA when `.grayscale()` is called were panicking with unsupported image type. This PR adds in the conversion for these types to Luma via the image crates automatic pixel conversion (likely removing the alpha channel).

Author: xd009642

.github/workflows/lints.yml

@@ -54,4 +54,3 @@ jobs:
         uses: actions-rs/cargo@v1
         with:
           command: clippy
-          args: -- -D warnings

akaze/src/image.rs

@@ -64,6 +64,26 @@ impl GrayFloatImage {
                     Luma([f32::from(gray_image[(x, y)][0]) / 65535f32])
                 })
             }
+            DynamicImage::ImageLumaA8(gray_image) => {
+                info!(
+                    "Loaded a {} x {} 8-bit image",
+                    input_image.width(),
+                    input_image.height()
+                );
+                ImageBuffer::from_fn(gray_image.width(), gray_image.height(), |x, y| {
+                    Luma([f32::from(gray_image[(x, y)][0]) / 255f32])
+                })
+            }
+            DynamicImage::ImageLumaA16(gray_image) => {
+                info!(
+                    "Loaded a {} x {} 16-bit image",
+                    input_image.width(),
+                    input_image.height()
+                );
+                ImageBuffer::from_fn(gray_image.width(), gray_image.height(), |x, y| {
+                    Luma([f32::from(gray_image[(x, y)][0]) / 65535f32])
+                })
+            }
             DynamicImage::ImageRgb32F(float_image) => {
                 info!(
                     "Loaded a {} x {} 32-bit RGB float image",

akaze/src/scale_space_extrema.rs

@@ -11,7 +11,7 @@ impl Akaze {
     /// # Argument
     /// * `evolutions` - evolutions to mutate in place.
     /// * `options` - options to use.
-    fn find_scale_space_extrema(&self, evolutions: &mut [EvolutionStep]) -> Vec<KeyPoint> {
+    fn find_scale_space_extrema(&self, evolutions: &[EvolutionStep]) -> Vec<KeyPoint> {
         let mut keypoint_cache: Vec<KeyPoint> = vec![];
         let smax = 10.0f32 * f32::sqrt(2.0f32);
         for (e_id, evolution) in evolutions.iter().enumerate() {
@@ -150,7 +150,7 @@ impl Akaze {
     /// * `options` - The options to use.
     /// # Return value
     /// The resulting keypoints.
-    pub fn detect_keypoints(&self, evolutions: &mut [EvolutionStep]) -> Vec<KeyPoint> {
+    pub fn detect_keypoints(&self, evolutions: &[EvolutionStep]) -> Vec<KeyPoint> {
         let mut keypoints = self.find_scale_space_extrema(evolutions);
         keypoints = do_subpixel_refinement(&keypoints, evolutions);
         keypoints