Android - How can I access the camera image

Android – How can I access the camera image

This article explains how to extract one of the images provided by the QCAR::Frame object, using a desired image format, and how to convey the image data from native C++ to Java (Android).

Taking the Image Targets sample code as a reference, you can follow these steps to achieve what is described above:

• In ImageTargets.cpp, add the following two global variables (javaVM and activityObj) and the function JNI_OnLoad (this function is automatically invoked and is used here for initializing the javaVM object, which represents the Java Virtual Machine of our Android Activity):

        //global variables

    JavaVM* javaVM = 0;

    jobject activityObj = 0;

    JNIEXPORT jint JNICALL

    JNI_OnLoad(JavaVM* vm,  void* reserved) {

        LOG("JNI_OnLoad");

        javaVM = vm;

        return JNI_VERSION_1_4;

    }

• Initialize the activityObj object by adding this code at the end of the initApplicationNative function:

    activityObj = env->NewGlobalRef(obj);

• Add the following code at the beginning of the QCAR_onUpdate function:

    QCAR::Image *imageRGB565 = NULL;

  QCAR::Frame frame = state.getFrame();

  for (int i = 0; i < frame.getNumImages(); ++i) {

    const QCAR::Image *image = frame.getImage(i);

    if (image->getFormat() == QCAR::RGB565) {

      imageRGB565 = (QCAR::Image*)image;

      break;

    }

  }

  if (imageRGB565) {

    JNIEnv* env = 0;

    if ((javaVM != 0) && (activityObj != 0) && (javaVM->GetEnv((void**)&env, JNI_VERSION_1_4) == JNI_OK)) {

      const short* pixels = (const short*) imageRGB565->getPixels();

      int width = imageRGB565->getWidth();

      int height = imageRGB565->getHeight();

      int numPixels = width * height;

      jbyteArray pixelArray = env->NewByteArray(numPixels * 2);

      env->SetByteArrayRegion(pixelArray, 0, numPixels * 2, (const jbyte*) pixels);

      jclass javaClass = env->GetObjectClass(activityObj);

      jmethodID method = env-> GetMethodID(javaClass, "processCameraImage", "([BII)V");

      env->CallObjectMethod(activityObj, method, pixelArray, width, height);

      env->DeleteLocalRef(pixelArray);

    }

  }

• Again in ImageTargets.cpp, in the startCamera function, add the code to specify the desired frame format, right after calling QCAR::CameraDevice::getInstance().start(), as shown in the following snippet:

    // Start the camera:

  if (!QCAR::CameraDevice::getInstance().start())

        return;

  QCAR::setFrameFormat(QCAR::RGB888, true);

  //  etc ...

• Finally, in ImageTargets.java, add a method called “processCameraImage” to process the byte array representing the image passed from JNI code; for instance, the following code uses the image buffer (byte[]) to build an Android Bitmap:

   public void processCameraImage(byte[] buffer, int width, int height)

 {

   Bitmap bitmap = Bitmap.createBitmap(width, height, Bitmap.Config.RGB_565);

   bitmap.copyPixelsFromBuffer(ByteBuffer.wrap(buffer));  

 }