Update README.
Check bit depth when saving image as PNG.

README.md

diff --git a/README.md b/README.md
index cd86f97..dc33e8c 100644
--- a/README.md
+++ b/README.md
@@ -3,11 +3,13 @@
 `TinyGLTF` is a header only C++11 glTF 2.0 https://github.com/KhronosGroup/glTF library.
 
 `TinyGLTF` uses Niels Lohmann's json library(https://github.com/nlohmann/json), so now it requires C++11 compiler.
-If you are looking for old, C++03 version, please use `devel-picojson` branch.  
+If you are looking for old, C++03 version, please use `devel-picojson` branch.
 
 ## Status
 
-v2.0.0 release(22 Aug, 2018)! 
+v2.2.0 release(Support loading 16bit PNG)
+v2.1.0 release(Draco support)
+v2.0.0 release(22 Aug, 2018)!
 
 ## Builds
 
@@ -38,14 +40,15 @@
 * Image(Using stb_image)
   * [x] Parse BASE64 encoded embedded image data(DataURI).
   * [x] Load external image file.
-  * [x] PNG(8bit only)
-  * [x] JPEG(8bit only)
-  * [x] BMP
-  * [x] GIF
+  * [x] Load PNG(8bit and 16bit)
+  * [x] Load JPEG(8bit only)
+  * [x] Load BMP
+  * [x] Load GIF
   * [x] Custom Image decoder callback(e.g. for decoding OpenEXR image)
 * Load from memory
 * Custom callback handler
   * [x] Image load
+  * [x] Image save
 * Extensions
   * [x] Draco mesh decoding
 
@@ -69,10 +72,12 @@
 * [ ] Mesh Compression/decompression(Open3DGC, etc)
   * [x] Load Draco compressed mesh
   * [x] Save Draco compressed mesh
+  * [ ] Open3DGC?
 * [ ] Support `extensions` and `extras` property
 * [ ] HDR image?
   * [ ] OpenEXR extension through TinyEXR.
-* [ ] Write example and tests for `animation` and `skin` 
+* [ ] 16bit PNG support in Serialization
+* [ ] Write example and tests for `animation` and `skin`
   * [ ] Skinning
   * [ ] Morph targets
 
@@ -104,13 +109,13 @@
 
 using namespace tinygltf;
 
-Model model; 
+Model model;
 TinyGLTF loader;
 std::string err;
 std::string warn;
-  
+
 bool ret = loader.LoadASCIIFromFile(&model, &err, &warn, argv[1]);
-//bool ret = loader.LoadBinaryFromFile(&model, &err, &warn, argv[1]); // for binary glTF(.glb) 
+//bool ret = loader.LoadBinaryFromFile(&model, &err, &warn, argv[1]); // for binary glTF(.glb)
 
 if (!warn.empty()) {
   printf("Warn: %s\n", warn.c_str());

tiny_gltf.h

diff --git a/tiny_gltf.h b/tiny_gltf.h
index 0338ab5..d675777 100644
--- a/tiny_gltf.h
+++ b/tiny_gltf.h
@@ -26,6 +26,7 @@
 // THE SOFTWARE.
 
 // Version:
+//  - v2.2.0 Add loading 16bit PNG support.
 //  - v2.1.0 Add draco compression.
 //  - v2.0.1 Add comparsion feature(Thanks to @Selmar).
 //  - v2.0.0 glTF 2.0!.
@@ -493,7 +494,7 @@
   int height;
   int component;
   int bits; // bit depth per channel. 8(byte), 16 or 32.
-  int pixel_type; // pixel type(TINYGLTF_COMPONENT_TYPE_***). usually UBYTE(bits = 8) or USHORT(bits = 16) 
+  int pixel_type; // pixel type(TINYGLTF_COMPONENT_TYPE_***). usually UBYTE(bits = 8) or USHORT(bits = 16)
   std::vector<unsigned char> image;
   int bufferView;        // (required if no uri)
   std::string mimeType;  // (required if no uri) ["image/jpeg", "image/png",
@@ -1772,6 +1773,11 @@
   std::vector<unsigned char> data;
 
   if (ext == "png") {
+    if ((image->bits != 8) || (image->pixel_type != TINYGLTF_COMPONENT_TYPE_UNSIGNED_BYTE)) {
+      // Unsupported pixel format
+      return false;
+    }
+
     if (!stbi_write_png_to_func(WriteToMemory_stbi, &data, image->width,
                                 image->height, image->component,
                                 &image->image[0], 0)) {
@@ -4905,7 +4911,7 @@
 
   // 12 bytes for header, JSON content length, 8 bytes for JSON chunk info, padding
   const int length = 12 + 8 + int(content.size()) + padding_size;
-  
+
   gltfFile.write(header.c_str(), header.size());
   gltfFile.write(reinterpret_cast<const char *>(&version), sizeof(version));
   gltfFile.write(reinterpret_cast<const char *>(&length), sizeof(length));