Hello,
I using a c++ dll provided by a customer that connects to a USB-video camera. The image captured by the camera uses OpenCV contructs such as IplImage. Since I need to create a Java applet to display the image provide by this .dll, I've create a JNI file that implements equivalent methods to the C++ code (see below). The problem with my C++ implementation is that I can't convert IplImage->imageData to a jbyteArray (it's always "null"). I'm not sure if I'm going about it the right way. I've used the JavaCV project but it's an overkill for my simple Java Applet. I would prefer to convert IplImage->imageData to some native Java structure instead. I can't figure out what is wrong with my C++ implementation. Suggestions?
MY JAVA MapleJNI Class:
public class MapleJNI {
static{
System.loadLibrary("MapleJNI.dll");
}
public static native byte maple_connect();
public static native void maple_disconnect();
public static native byte maple_set_capture_mode(byte mode);
public static native byte maple_capture_image(byte image_type, byte[] img_buf);
public static native void maple_crop_image(byte[] img_buf);
[ ... ]
C++ - Code ".cpp" File (JNI Implementation)
JNIEXPORT jbyte JNICALL Java_maple_MapleJNI_maple_1capture_1image (JNIEnv *env, jclass, jbyte mode, jbyteArray image){
UCHAR result = maple_capture_image((UCHAR) mode, (UCHAR *) rgb_img->imageData);
jsize len = 0;
if (result == STATUS_NO_ERROR)
{
// Crop top/bot rows of image
// (set to black) to avoid jitter
maple_crop_image((UCHAR *) rgb_img->imageData);
len = sizeof(rgb_img->imageData);
image = env->NewByteArray(len);
// Copy "rgb_image" to java native jbyteArray "image"
env->SetByteArrayRegion(image, 0, len, (jbyte*) rgb_img->imageData);
}
else{
maple_disconnect();
}
return result;
}