美文网首页androidAndroid开发Android知识
Android Studio 2.2 JNI编译及Rxjava使

Android Studio 2.2 JNI编译及Rxjava使

作者: nothingwxq | 来源:发表于2016-09-28 20:46 被阅读578次
    jni-2.png

    背景

    最近几天刚好无事由于在新公司,业务上安排不是很满。android studio 2.2以后,jni比较方便开发了。本文是使用jni进行初级的demo需求。一个图片的高斯模糊效果。算法参见(https://github.com/GankLi/Demo/tree/demo/app/src/main/java/com/gank/demo/gaussblurtest)。 上图是未模糊前原图,恩,我会把它模糊滴。

    前提

    请下下好ndk和cmake工具。需要环境android studio 2.2

    down.jpeg

    快速入门

    总的工程图如图:

    总的框架.png

    jni配置想具体了解的可以看官方,本文仅提供一个简要的模式。先上gradle配置

    apply plugin: 'com.android.application'
    
    android {
        compileSdkVersion 23
        buildToolsVersion '23.0.2'
        defaultConfig {
            applicationId 'com.example.hellojni'
            minSdkVersion 15
            targetSdkVersion 23
            versionCode 1
            versionName "1.0"
            externalNativeBuild {
                cmake {
                    arguments '-DANDROID_TOOLCHAIN=clang'
                }
            }
        }
        buildTypes {
            release {
                minifyEnabled false
                proguardFiles getDefaultProguardFile('proguard-android.txt'), 'proguard-rules.pro'
            }
        }
        externalNativeBuild {
            cmake {
                path "src/main/cpp/CMakeLists.txt"
            }
        }
    }
    
    dependencies {
        compile fileTree(dir: 'libs', include: ['*.jar'])
        compile 'com.android.support:appcompat-v7:23.4.0'
        compile 'com.github.bumptech.glide:glide:3.7.0'
        compile 'io.reactivex:rxjava:1.1.0'
        compile 'io.reactivex:rxandroid:1.1.0'
    }
    

    其中jni相关的

    externalNativeBuild {
                cmake {
                    arguments '-DANDROID_TOOLCHAIN=clang'
                }
            }
    

    cmake,这里指明了clang来作为编译器,这里其他配置参见官方。

     externalNativeBuild {
            cmake {
                path "src/main/cpp/CMakeLists.txt"
            }
        }
    

    这里指明make file相关的编译文件路径。

    BlurUtil.java

    package com.example.nothing.blurdemo;
    
    import android.graphics.Bitmap;
    
    public class BlurUtil {
        //分别在x轴 和 y轴方向上进行高斯模糊
        public static Bitmap gaussBlurUseGauss(Bitmap bitmap, int radius) {
            int w = bitmap.getWidth();
            int h = bitmap.getHeight();
    
            //生成一张新的图片
            Bitmap outBitmap = Bitmap.createBitmap(w, h, Bitmap.Config.ARGB_8888);
    
            //定义一个临时数组存储原始图片的像素 值
            int[] pix = new int[w * h];
    
            //将图片像素值写入数组
            bitmap.getPixels(pix, 0, w, 0, 0, w, h);
    
            //进行模糊
            initCBlur1(pix, w, h, radius);
    
            //将数据写入到 图片
            outBitmap.setPixels(pix, 0, w, 0, 0, w, h);
    
            //返回结果
            return outBitmap;
        }
    
        //利用均值模糊 逼近 高斯模糊
        public static Bitmap gaussBlurUseAvg(Bitmap bitmap, int radius) {
            int w = bitmap.getWidth();
            int h = bitmap.getHeight();
    
            //生成一张新的图片
            Bitmap outBitmap = Bitmap.createBitmap(w, h, Bitmap.Config.ARGB_8888);
    
            //定义一个临时数组存储原始图片的像素 值
            int[] pix = new int[w * h];
    
            //将图片像素值写入数组
            bitmap.getPixels(pix, 0, w, 0, 0, w, h);
    
            //进行模糊
            initCBlur2(pix, w, h, radius);
    
            //将数据写入到 图片
            outBitmap.setPixels(pix, 0, w, 0, 0, w, h);
    
            //返回结果
            return outBitmap;
        }
    
        //原始的高斯模糊 方法
        private static native void initCBlur1(int[] pix, int w, int h, int r);
    
        //利用均值模糊进行拟合 高斯模糊
        private static native void initCBlur2(int[] pix, int w, int h, int r);
    
        //加载native模块
        static {
            System.loadLibrary("hello-jni");
        }
    }
    

    主要注意加载模块。

    hello-jni.cpp jni文件

    由于高斯模糊,代码不多,就写到同一个cpp文件中。

    #include <jni.h>
    #include <android/log.h>
    #include <iostream>
    #include <cmath>
    
    #define  LOG_TAG    "blur"
    #define  LOGD(...)  __android_log_print(ANDROID_LOG_DEBUG, LOG_TAG, __VA_ARGS__)
    
    #define PI 3.1415926
    
    
    extern "C" {
    
    //LOGD("jni %d:  %lf", i, amplitude[i]);
    void gaussBlur1(int *pix, int w, int h, int radius) {
        float sigma = 1.0 * radius / 2.57;    //2.57 * sigam半径之后基本没有贡献 所以取sigma为 r / 2.57
        float deno = 1.0 / (sigma * sqrt(2.0 * PI));
        float nume = -1.0 / (2.0 * sigma * sigma);
    
        //高斯分布产生的数组
        float *gaussMatrix = (float *) malloc(sizeof(float) * (radius + radius + 1));
        float gaussSum = 0.0;
        for (int i = 0, x = -radius; x <= radius; ++x, ++i) {
            float g = deno * exp(1.0 * nume * x * x);
    
            gaussMatrix[i] = g;
            gaussSum += g;
        }
    
        //归1话
        int len = radius + radius + 1;
        for (int i = 0; i < len; ++i)
            gaussMatrix[i] /= gaussSum;
    
        //临时存储 一行的数据
        int *rowData = (int *) malloc(w * sizeof(int));
        int *listData = (int *) malloc(h * sizeof(int));
    
        //x方向的模糊
        for (int y = 0; y < h; ++y) {
            //拷贝一行数据
            memcpy(rowData, pix + y * w, sizeof(int) * w);
    
            for (int x = 0; x < w; ++x) {
                float r = 0, g = 0, b = 0;
                gaussSum = 0;
    
                for (int i = -radius; i <= radius; ++i) {
                    int k = x + i;
    
                    if (0 <= k && k <= w) {
                        //得到像素点的rgb值
                        int color = rowData[k];
                        int cr = (color & 0x00ff0000) >> 16;
                        int cg = (color & 0x0000ff00) >> 8;
                        int cb = (color & 0x000000ff);
    
                        r += cr * gaussMatrix[i + radius];
                        g += cg * gaussMatrix[i + radius];
                        b += cb * gaussMatrix[i + radius];
    
                        gaussSum += gaussMatrix[i + radius];
                    }
                }
    
                int cr = (int) (r / gaussSum);
                int cg = (int) (g / gaussSum);
                int cb = (int) (b / gaussSum);
    
                pix[y * w + x] = cr << 16 | cg << 8 | cb | 0xff000000;
            }
        }
    
        for (int x = 0; x < w; ++x) {
            //拷贝 一列 数据
            for (int y = 0; y < h; ++y)
                listData[y] = pix[y * w + x];
    
            for (int y = 0; y < h; ++y) {
                float r = 0, g = 0, b = 0;
                gaussSum = 0;
    
                for (int j = -radius; j <= radius; ++j) {
                    int k = y + j;
    
                    if (0 <= k && k <= h) {
                        int color = listData[k];
                        int cr = (color & 0x00ff0000) >> 16;
                        int cg = (color & 0x0000ff00) >> 8;
                        int cb = (color & 0x000000ff);
    
                        r += cr * gaussMatrix[j + radius];
                        g += cg * gaussMatrix[j + radius];
                        b += cb * gaussMatrix[j + radius];
    
                        gaussSum += gaussMatrix[j + radius];
                    }
                }
    
                int cr = (int) (r / gaussSum);
                int cg = (int) (g / gaussSum);
                int cb = (int) (b / gaussSum);
    
                pix[y * w + x] = cr << 16 | cg << 8 | cb | 0xff000000;
            }
        }
    
        //清理内存
        free(gaussMatrix);
        free(rowData);
        free(listData);
    }
    
    
    //参考:http://blog.ivank.net/fastest-gaussian-blur.html
    //横向的均值模糊 srcPix:原始的像素值 destPix将处理过的像素值放入到 destPix中
    void boxBlurH(int *srcPix, int *destPix, int w, int h, int radius) {
        //用于索引
        int index;
    
        //r g b在遍历是 累加的色彩通道的总和
        int a = 0, r = 0, g = 0, b = 0;
        int ta, tr, tg, tb;    //临时变量
    
        //临时变量
        int color;
        int preColor;
    
        //用于计算权值 1 / num
        int num;
        float iarr;
    
        for (int i = 0; i < h; ++i) {
            r = 0;
            g = 0;
            b = 0;
    
            index = i * w;
            num = radius;
    
            for (int j = 0; j < radius; j++) {
                //累加0,radius-1的色彩的总和
                color = srcPix[index + j];
                //a += (color & 0xff000000) >> 24;
                r += (color & 0x00ff0000) >> 16;
                g += (color & 0x0000ff00) >> 8;
                b += (color & 0x000000ff);
            }
    
            //真正开始计算
            for (int j = 0; j <= radius; ++j) {
                num++;
                iarr = 1.0 / (1.0 * num);
    
                color = srcPix[index + j + radius];
                //a += (color & 0xff000000) >> 24;
                r += (color & 0x00ff0000) >> 16;
                g += (color & 0x0000ff00) >> 8;
                b += (color & 0x000000ff);
    
                //ta = (int)(1.0 * a / num);
                tr = (int) (r * iarr);
                tg = (int) (g * iarr);
                tb = (int) (b * iarr);
    
                destPix[index + j] = tr << 16 | tg << 8 | tb | 0xff000000;
            }
    
            iarr = 1.0 / (1.0 * num);
            for (int j = radius + 1; j < w - radius; ++j) {
                preColor = srcPix[index + j - 1 - radius];
                color = srcPix[index + j + radius];
    
                //a += (color & 0xff000000) >> 24 - (preColor & 0xff000000) >> 24;
                r = r + ((color & 0x00ff0000) >> 16) - ((preColor & 0x00ff0000) >> 16);
                g = g + ((color & 0x0000ff00) >> 8) - ((preColor & 0x0000ff00) >> 8);
                b = b + (color & 0x000000ff) - (preColor & 0x000000ff);
    
                //ta = (int)(1.0 * a / num);
                tr = (int) (r * iarr);
                tg = (int) (g * iarr);
                tb = (int) (b * iarr);
    
                destPix[index + j] = tr << 16 | tg << 8 | tb | 0xff000000;
            }
    
            for (int j = w - radius; j < w; ++j) {
                num--;
                iarr = 1.0 / (1.0 * num);
    
                preColor = srcPix[index + j - 1 - radius];
    
                //a -= (preColor & 0xff000000) >> 24;
                r -= (preColor & 0x00ff0000) >> 16;
                g -= (preColor & 0x0000ff00) >> 8;
                b -= (preColor & 0x000000ff);
    
                //ta = (int)(1.0 * a / num);
                tr = (int) (r * iarr);
                tg = (int) (g * iarr);
                tb = (int) (b * iarr);
    
                //
                //destPix[index + j] = (ta << 24 | tr << 16 | tg << 8 | tb);
                destPix[index + j] = tr << 16 | tg << 8 | tb | 0xff000000;
            }
        }
    }
    
    
    //列的均值模糊 srcPix:原始的像素值 destPix将处理过的像素值放入到 destPix中
    void boxBlurV(int *srcPix, int *destPix, int w, int h, int radius) {
        //r g b在遍历是 累加的色彩通道的总和
        int a = 0, r = 0, g = 0, b = 0;
        int ta, tr, tg, tb;    //临时变量
    
        //临时变量
        int color;
        int preColor;
    
        //用于计算权值 1 / num
        int num;
        float iarr;
    
        for (int i = 0; i < w; ++i) {
            r = 0;
            g = 0;
            b = 0;
    
            num = radius;
    
            for (int j = 0; j < radius; ++j) {
                color = srcPix[j * w + i];
                r += (color & 0x00ff0000) >> 16;
                g += (color & 0x0000ff00) >> 8;
                b += (color & 0x000000ff);
            }
    
            for (int j = 0; j <= radius; ++j) {
                num++;
                iarr = 1.0 / (1.0 * num);
    
                color = srcPix[(j + radius) * w + i];
                r += (color & 0x00ff0000) >> 16;
                g += (color & 0x0000ff00) >> 8;
                b += (color & 0x000000ff);
    
                tr = (int) (r * iarr);
                tg = (int) (g * iarr);
                tb = (int) (b * iarr);
    
                destPix[j * w + i] = tr << 16 | tg << 8 | tb | 0xff000000;
            }
    
            iarr = 1.0 / (1.0 * num);
            for (int j = radius + 1; j < h - radius; ++j) {
                preColor = srcPix[(j - radius - 1) * w + i];
                color = srcPix[(j + radius) * w + i];
    
                r = r + ((color & 0x00ff0000) >> 16) - ((preColor & 0x00ff0000) >> 16);
                g = g + ((color & 0x0000ff00) >> 8) - ((preColor & 0x0000ff00) >> 8);
                b = b + (color & 0x000000ff) - (preColor & 0x000000ff);
    
                tr = (int) (r * iarr);
                tg = (int) (g * iarr);
                tb = (int) (b * iarr);
    
                destPix[j * w + i] = tr << 16 | tg << 8 | tb | 0xff000000;
            }
    
            for (int j = h - radius; j < h; ++j) {
                num--;
                iarr = 1.0 / (1.0 * num);
                preColor = srcPix[(j - radius - 1) * w + i];
    
                r -= (preColor & 0x00ff0000) >> 16;
                g -= (preColor & 0x0000ff00) >> 8;
                b -= (preColor & 0x000000ff);
    
                tr = (int) (r * iarr);
                tg = (int) (g * iarr);
                tb = (int) (b * iarr);
    
                destPix[j * w + i] = tr << 16 | tg << 8 | tb | 0xff000000;
            }
        }
    }
    
    void boxBlur(int *srcPix, int *destPix, int w, int h, int r) {
        if (r < 0) {
            LOGD("boxBlur r < 0: %d", r);
            return;
        }
    
        boxBlurH(srcPix, destPix, w, h, r);
        boxBlurV(destPix, srcPix, w, h, r);
    }
    
    //领用n 个 box 拟合 sigma的高斯函数
    //参考:http://www.csse.uwa.edu.au/~pk/research/pkpapers/FastGaussianSmoothing.pdf
    void boxesForGauss(float sigma, int *size, int n) {
        float wIdeal = sqrt(12.0 * sigma * sigma / n + 1.0);
        int wl = floor(wIdeal);
    
        if (0 == wl % 2)
            wl--;
    
        int wu = wl + 2;
    
        float mIdeal = (12.0 * sigma * sigma - n * wl * wl - 4 * n * wl - 3 * n) / (-4 * wl - 4);
        int m = round(mIdeal);
    
        for (int i = 0; i < n; ++i)
            size[i] = (i < m ? wl : wu);
    }
    
    void gaussBlur2(int *pix, int w, int h, int r) {
        float sigma = 1.0 * r / 2.57;    //2.57 *sigam半径之后基本没有贡献 所以取sigma为 r / 2.57
    
        int boxSize = 3;
        int *boxR = (int *) malloc(sizeof(int) * boxSize);    //需要的个数
    
        //计算拟合的半径
        boxesForGauss(sigma, boxR, boxSize);
    
        int *tempPix = (int *) malloc(sizeof(int) * w * h);
    
        boxBlur(pix, tempPix, w, h, (boxR[0] - 1) / 2);
        boxBlur(pix, tempPix, w, h, (boxR[1] - 1) / 2);
        boxBlur(pix, tempPix, w, h, (boxR[2] - 1) / 2);
    
        //清理内存
        free(boxR);
        free(tempPix);
    }
    
    void Java_com_example_nothing_blurdemo_BlurUtil_initCBlur1(JNIEnv *env,
                                                               jobject obj,
                                                               jintArray pix,
                                                               jint w,
                                                               jint h,
                                                               jint r) {
        gaussBlur1(env->GetIntArrayElements(pix, NULL), w, h, r);
    }
    
    void Java_com_example_nothing_blurdemo_BlurUtil_initCBlur2(JNIEnv *env,
                                                               jobject obj,
                                                               jintArray pix,
                                                               jint w,
                                                               jint h,
                                                               jint r) {
        gaussBlur2(env->GetIntArrayElements(pix, NULL), w, h, r);
    }
    
    }
    

    注意这里:
    Java_com_example_nothing_blurdemo_BlurUtil 为BlurUtil类的全包路径。initCBlur1 方法名。JNIEnv 环境变量,jobject ,jintArray什么的,j代表java,去掉j就知道了。

    void Java_com_example_nothing_blurdemo_BlurUtil_initCBlur1(JNIEnv *env,
                                                               jobject obj,
                                                               jintArray pix,
                                                               jint w,
                                                               jint h,
                                                               jint r) {
        gaussBlur1(env->GetIntArrayElements(pix, NULL), w, h, r);
    }
    

    CMakeLists.txt文件

    cmake_minimum_required(VERSION 3.4.1)
    
    add_library(hello-jni SHARED
                hello-jni.cpp)
    
    # Include libraries needed for hello-jni lib
    target_link_libraries(hello-jni log android)
    

    add_library(hello-jni SHARED
    hello-jni.cpp) 三个参数 ,第一个为库名,第二个shared库(so)。最后一个参数是引入的文件。

    MainActivity

    public class MainActivity extends AppCompatActivity {
    
        private static final String TAG = "nothingwxq";
        ImageView mImageView;
        private Subscriber<Bitmap> mBitmapSubscriber;
        private Subscription mSubscription;
    
        @Override
        protected void onCreate(Bundle savedInstanceState) {
            super.onCreate(savedInstanceState);
            setContentView(R.layout.activity_main);
            mImageView = (ImageView) this.findViewById(R.id.image);
    
            mBitmapSubscriber = new Subscriber<Bitmap>() {
                @Override
                public void onCompleted() {
                    Log.d(TAG,"onCompleted---");
                }
    
                @Override
                public void onError(Throwable e) {
    
                }
    
                @Override
                public void onNext(Bitmap bitmap) {
                    Log.d(TAG,"onNext---");
                    mImageView.setImageBitmap(bitmap);
                }
            };
    
            Glide.with(getApplicationContext())
                    .load("http://ww2.sinaimg.cn/mw1024/005yr3jYjw1f23h3vg4waj30qo0ziwlx.jpg")
                    .asBitmap()
                    .into(new SimpleTarget<Bitmap>() {
                        @Override
                        public void onResourceReady(final Bitmap bitmap, GlideAnimation anim) {
                            mSubscription = Observable
                                    .create(new Observable.OnSubscribe<Bitmap>() {
                                        @Override
                                        public void call(Subscriber<? super Bitmap> subscriber) {
                                            Log.d(TAG, "creat---");
                                            subscriber.onNext(bitmap);
                                            subscriber.onCompleted();
                                        }
                                    })
                                    .observeOn(Schedulers.io())
                                    .map(new Func1<Bitmap, Bitmap>() {
                                        @Override
                                        public Bitmap call(Bitmap bitmap) {
                                            Log.d(TAG, "map---");
                                            return BlurUtil.gaussBlurUseAvg(bitmap, 0);
                                        }
                                    })
                                    .subscribeOn(AndroidSchedulers.mainThread())
                                    .observeOn(AndroidSchedulers.mainThread())
                                    .subscribe(mBitmapSubscriber);
                        }
                    });
    
        }
    
        @Override
        protected void onDestroy() {
            super.onDestroy();
            mSubscription.unsubscribe();
        }
    }
    

    本文省略资源文件,很简单的,一张图片。这里说下逻辑,使用Glide加载网上的图片,这里涉及Glide回调的监听。下载完成后,通过rxjava 的map 操作处理,最后onNext中进行设置图片,ok,完工。看下效果:


    jni-3.png

    相关文章

      网友评论

      本文标题:Android Studio 2.2 JNI编译及Rxjava使

      本文链接:https://www.haomeiwen.com/subject/cbckyttx.html