Certiface e a tecnologia Intel no combate a fraude.
-
Upload
alessandro-faria -
Category
Technology
-
view
227 -
download
0
description
Transcript of Certiface e a tecnologia Intel no combate a fraude.
Alessandro de Oliveira Faria (A.K.A. Cabelo)
Certiface e a tecnologia Intel no combate a Fraude.
ApresentaçãoApresentação
Ingressouse aos 12 anos no mercado de tecnologia/inovação.
● Sócio fundador da NETi TECNOLOGIA;● Inventor da tecnologia Certiface;● 102 Artigos publicados na internet (Viva O Linux e outros);● Embaixador openSUSE na América Latina;● Experiência com Visão Computacional (openCV e GPU );
Fraudadores: Briga de gato e rato!Fraudadores: Briga de gato e rato!
SUPER
99,95%
O Certiface: Protege as pessoas O Certiface: Protege as pessoas
O Certiface: Obrigado Intel... O Certiface: Obrigado Intel...
Certiface: Nasceu grande... Certiface: Nasceu grande...
PCL : Point Cloud LibraryPCL : Point Cloud Library
$ unzip androidndkr4blinuxx86.zip
Pronto o NDK esta instalado com sucesso!
Instalação e configuração NDK:Instalação e configuração NDK:
public class HelloNDK extends Activity { TextView txtHello;
@Override public void onCreate(Bundle savedInstanceState) { super.onCreate(savedInstanceState); setContentView(R.layout.main);
txtHello = (TextView) findViewById(R.id.txtHello); String hello = sayHello(); txtHello.setText(hello); }
public native String sayHello(); static { System.loadLibrary("hellondk"); }}
$ cd ~/workspace/HelloNDK$ javah o hellondk.h classpath bin com.cabelo.hellondk.HelloNDK$ mkdir jni$ cp hellondk.h jni/$ cd jni
$ vi hellondk.cpp
#include <hellondk.h>
JNIEXPORT jstring JNICALL Java_com_cabelo_hellondk_HelloNDK_sayHello(JNIEnv *env, jobject obj){ return env>NewStringUTF("Hello, NDK!"); }
PictureCallback rawCallback = new PictureCallback() {public void onPictureTaken(byte[] _data, Camera _camera) {[ … ] } };
voidJava_com_cabelo_GetFace_PreviewView_RgbToGray( JNIEnv* env, jobject this, jbyteArray src, jintArray dst ){
jboolean b;jint len;jbyte* srcPtr=(*env)>GetByteArrayElements(env,src,&b);jint* dstPtr=(*env)>GetIntArrayElements(env,dst,&b);len = (*env)>GetArrayLength(env,src);RGB32toGray8(srcPtr, dstPtr, len);(*env)>ReleaseByteArrayElements(env, src, srcPtr, 0);(*env)>ReleaseIntArrayElements(env, dst, dstPtr, 0);
};
$ vi Application.mk
APP_ABI := armeabi armeabiv7a x86
$ /usr/local/share/androidndkr4b/ndkbuild APP_ABI=all Compile++ thumb: hellondk <= ~/HelloNDK/jni/hellondk.cpp SharedLibrary : libhellondk.so Install : libhellondk.so => ~/HelloNDK/libs/armeabi Compile++ thumb: hellondk <= ~/HelloNDK/jni/hellondk.cpp SharedLibrary : libhellondk.so Install : libhellondk.so => ~/HelloNDK/libs/armeabiv7 Compile++ thumb: hellondk <= ~/HelloNDK/jni/hellondk.cpp SharedLibrary : libhellondk.so Install : libhellondk.so => ~/HelloNDK/libs/x86
OpenCV for AndroidOpenCV for Android
10 x mais performático.http://docs.opencv.org/doc/tutorials/introduction/android_binary_package/O4A_SDK.html
https://www.threadingbuildingblocks.org/
https://software.intel.com/en-us/articles/using-intel-ipp-with-android-os
gmake tbb tbbmalloc target=android
https://software.intel.com/en-us/android/articles/android-tutorial-writing-a-multithreaded-application-using-intel-threading-building-blocks
TBB e IPP for AndroidTBB e IPP for Android
FaceDetector face_detector = new FaceDetector( background_image.getWidth(), background_image.getHeight(), MAX_FACES);
faces = new FaceDetector.Face[MAX_FACES];face_count = face_detector.findFaces(background_image, faces);Log.d("Face_Detection", "Face Count: " + String.valueOf(face_count));
https://software.intel.com/en-us/blogs/2013/10/28/implementing-face-detection-in-android
Detectar facesDetectar faces
Obrigado...