caffe实现上采样知识点总结_上采样算法carafe

线性差值-反卷积-空洞卷积-上采样

等等一大堆名称，梳理梳理：

caffe中的deconvolution和upsample的区别？

https://www.zhihu.com/question/63890195

Caffe实现上采样（upsample）方法总结

https://blog.csdn.net/chaipp0607/article/details/95599218

我也问了作者一个问题

关于：

https://github.com/TimoSaemann/caffe-segnet-cudnn5/blob/master/src/caffe/layers/upsample_layer.cpp

和caffe-yolov3实现的upsample_layer.cpp有什么区别？？？

两个代码是不一样

 caffe-yolov3中的：

https://pan.baidu.com/share/init?surl=3GpoYoqKSCeFX0m0ves_fQ

密码bwrd

---

在Caffe下自定义网络层 Interp层

https://blog.csdn.net/donkey_1993/article/details/81180059

caffe系列：deeplab中的插值网络层前传和反传的实现分析

https://blog.csdn.net/xizero00/article/details/74330652

Caffe中Interp层的使用

https://www.cnblogs.com/wmr95/p/8715607.html

 

 

https://github.com/hszhao/PSPNet 

 

---

pytorch中的上采样：

https://blog.csdn.net/qq_31622015/article/details/90573618

 

Pytorch - torch.nn.modules.upsampling 和 interpolate 函数

https://www.aiuai.cn/aifarm605.html

---

关于yolov3的上采样：

https://blog.csdn.net/qq_34199326/article/details/84072505

原始的darknet：

https://blog.csdn.net/Lin_Danny/article/details/86514126

#include "upsample_layer.h"
#include "cuda.h"
#include "blas.h"
 
#include <stdio.h>
 
layer make_upsample_layer(int batch, int w, int h, int c, int stride)
{
    layer l = {0};
    l.type = UPSAMPLE;
    l.batch = batch;
    l.w = w;
    l.h = h;
    l.c = c;
    l.out_w = w*stride;
    l.out_h = h*stride;
    l.out_c = c;
    if(stride < 0){
        stride = -stride;
        l.reverse=1;
        l.out_w = w/stride;
        l.out_h = h/stride;
    }
    l.stride = stride;
    l.outputs = l.out_w*l.out_h*l.out_c;
    l.inputs = l.w*l.h*l.c;
    l.delta =  calloc(l.outputs*batch, sizeof(float));
    l.output = calloc(l.outputs*batch, sizeof(float));;
 
    l.forward = forward_upsample_layer;
    l.backward = backward_upsample_layer;
    #ifdef GPU
    l.forward_gpu = forward_upsample_layer_gpu;
    l.backward_gpu = backward_upsample_layer_gpu;
 
    l.delta_gpu =  cuda_make_array(l.delta, l.outputs*batch);
    l.output_gpu = cuda_make_array(l.output, l.outputs*batch);
    #endif
    if(l.reverse) fprintf(stderr, "downsample         %2dx  %4d x%4d x%4d   ->  %4d x%4d x%4d\n", stride, w, h, c, l.out_w, l.out_h, l.out_c);
    else fprintf(stderr, "upsample           %2dx  %4d x%4d x%4d   ->  %4d x%4d x%4d\n", stride, w, h, c, l.out_w, l.out_h, l.out_c);
    return l;
}
 
void resize_upsample_layer(layer *l, int w, int h)
{
    l->w = w;
    l->h = h;
    l->out_w = w*l->stride;
    l->out_h = h*l->stride;
    if(l->reverse){
        l->out_w = w/l->stride;
        l->out_h = h/l->stride;
    }
    l->outputs = l->out_w*l->out_h*l->out_c;
    l->inputs = l->h*l->w*l->c;
    l->delta =  realloc(l->delta, l->outputs*l->batch*sizeof(float));
    l->output = realloc(l->output, l->outputs*l->batch*sizeof(float));
 
#ifdef GPU
    cuda_free(l->output_gpu);
    cuda_free(l->delta_gpu);
    l->output_gpu  = cuda_make_array(l->output, l->outputs*l->batch);
    l->delta_gpu   = cuda_make_array(l->delta,  l->outputs*l->batch);
#endif
    
}
 
void forward_upsample_layer(const layer l, network net)
{
    fill_cpu(l.outputs*l.batch, 0, l.output, 1);
    if(l.reverse){
        upsample_cpu(l.output, l.out_w, l.out_h, l.c, l.batch, l.stride, 0, l.scale, net.input);
    }else{
        upsample_cpu(net.input, l.w, l.h, l.c, l.batch, l.stride, 1, l.scale, l.output);
    }
}
 
void backward_upsample_layer(const layer l, network net)
{
    if(l.reverse){
        upsample_cpu(l.delta, l.out_w, l.out_h, l.c, l.batch, l.stride, 1, l.scale, net.delta);
    }else{
        upsample_cpu(net.delta, l.w, l.h, l.c, l.batch, l.stride, 0, l.scale, l.delta);
    }
}
 
#ifdef GPU
void forward_upsample_layer_gpu(const layer l, network net)
{
    fill_gpu(l.outputs*l.batch, 0, l.output_gpu, 1);
    if(l.reverse){
        upsample_gpu(l.output_gpu, l.out_w, l.out_h, l.c, l.batch, l.stride, 0, l.scale, net.input_gpu);
    }else{
        upsample_gpu(net.input_gpu, l.w, l.h, l.c, l.batch, l.stride, 1, l.scale, l.output_gpu);
    }
}
 
void backward_upsample_layer_gpu(const layer l, network net)
{
    if(l.reverse){
        upsample_gpu(l.delta_gpu, l.out_w, l.out_h, l.c, l.batch, l.stride, 1, l.scale, net.delta_gpu);
    }else{
        upsample_gpu(net.delta_gpu, l.w, l.h, l.c, l.batch, l.stride, 0, l.scale, l.delta_gpu);
    }
}
#endif

其中：

void upsample_cpu(float *in, int w, int h, int c, int batch, int stride, int forward, float scale, float *out)
{
    int i, j, k, b;
    for(b = 0; b < batch; ++b){
        for(k = 0; k < c; ++k){
            for(j = 0; j < h*stride; ++j){
                for(i = 0; i < w*stride; ++i){
                    int in_index = b*w*h*c + k*w*h + (j/stride)*w + i/stride;
                    int out_index = b*w*h*c*stride*stride + k*w*h*stride*stride + j*w*stride + i;
                    if(forward) out[out_index] = scale*in[in_index];
                    else in[in_index] += scale*out[out_index];
                }
            }
        }
    }
}

 

 来自：

#include "blas.h"
 
#include <math.h>
#include <assert.h>
#include <float.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
void reorg_cpu(float *x, int w, int h, int c, int batch, int stride, int forward, float *out)
{
    int b,i,j,k;
    int out_c = c/(stride*stride);
 
    for(b = 0; b < batch; ++b){
        for(k = 0; k < c; ++k){
            for(j = 0; j < h; ++j){
                for(i = 0; i < w; ++i){
                    int in_index  = i + w*(j + h*(k + c*b));
                    int c2 = k % out_c;
                    int offset = k / out_c;
                    int w2 = i*stride + offset % stride;
                    int h2 = j*stride + offset / stride;
                    int out_index = w2 + w*stride*(h2 + h*stride*(c2 + out_c*b));
                    if(forward) out[out_index] = x[in_index];
                    else out[in_index] = x[out_index];
                }
            }
        }
    }
}
 
void flatten(float *x, int size, int layers, int batch, int forward)
{
    float *swap = calloc(size*layers*batch, sizeof(float));
    int i,c,b;
    for(b = 0; b < batch; ++b){
        for(c = 0; c < layers; ++c){
            for(i = 0; i < size; ++i){
                int i1 = b*layers*size + c*size + i;
                int i2 = b*layers*size + i*layers + c;
                if (forward) swap[i2] = x[i1];
                else swap[i1] = x[i2];
            }
        }
    }
    memcpy(x, swap, size*layers*batch*sizeof(float));
    free(swap);
}
 
void weighted_sum_cpu(float *a, float *b, float *s, int n, float *c)
{
    int i;
    for(i = 0; i < n; ++i){
        c[i] = s[i]*a[i] + (1-s[i])*(b ? b[i] : 0);
    }
}
 
void weighted_delta_cpu(float *a, float *b, float *s, float *da, float *db, float *ds, int n, float *dc)
{
    int i;
    for(i = 0; i < n; ++i){
        if(da) da[i] += dc[i] * s[i];
        if(db) db[i] += dc[i] * (1-s[i]);
        ds[i] += dc[i] * (a[i] - b[i]);
    }
}
 
void shortcut_cpu(int batch, int w1, int h1, int c1, float *add, int w2, int h2, int c2, float s1, float s2, float *out)
{
    int stride = w1/w2;
    int sample = w2/w1;
    assert(stride == h1/h2);
    assert(sample == h2/h1);
    if(stride < 1) stride = 1;
    if(sample < 1) sample = 1;
    int minw = (w1 < w2) ? w1 : w2;
    int minh = (h1 < h2) ? h1 : h2;
    int minc = (c1 < c2) ? c1 : c2;
 
    int i,j,k,b;
    for(b = 0; b < batch; ++b){
        for(k = 0; k < minc; ++k){
            for(j = 0; j < minh; ++j){
                for(i = 0; i < minw; ++i){
                    int out_index = i*sample + w2*(j*sample + h2*(k + c2*b));
                    int add_index = i*stride + w1*(j*stride + h1*(k + c1*b));
                    out[out_index] = s1*out[out_index] + s2*add[add_index];
                }
            }
        }
    }
}
 
void mean_cpu(float *x, int batch, int filters, int spatial, float *mean)
{
    float scale = 1./(batch * spatial);
    int i,j,k;
    for(i = 0; i < filters; ++i){
        mean[i] = 0;
        for(j = 0; j < batch; ++j){
            for(k = 0; k < spatial; ++k){
                int index = j*filters*spatial + i*spatial + k;
                mean[i] += x[index];
            }
        }
        mean[i] *= scale;
    }
}
 
void variance_cpu(float *x, float *mean, int batch, int filters, int spatial, float *variance)
{
    float scale = 1./(batch * spatial - 1);
    int i,j,k;
    for(i = 0; i < filters; ++i){
        variance[i] = 0;
        for(j = 0; j < batch; ++j){
            for(k = 0; k < spatial; ++k){
                int index = j*filters*spatial + i*spatial + k;
                variance[i] += pow((x[index] - mean[i]), 2);
            }
        }
        variance[i] *= scale;
    }
}
 
void l2normalize_cpu(float *x, float *dx, int batch, int filters, int spatial)
{
    int b,f,i;
    for(b = 0; b < batch; ++b){
        for(i = 0; i < spatial; ++i){
            float sum = 0;
            for(f = 0; f < filters; ++f){
                int index = b*filters*spatial + f*spatial + i;
                sum += powf(x[index], 2);
            }
            sum = sqrtf(sum);
            for(f = 0; f < filters; ++f){
                int index = b*filters*spatial + f*spatial + i;
                x[index] /= sum;
                dx[index] = (1 - x[index]) / sum;
            }
        }
    }
}
 
 
void normalize_cpu(float *x, float *mean, float *variance, int batch, int filters, int spatial)
{
    int b, f, i;
    for(b = 0; b < batch; ++b){
        for(f = 0; f < filters; ++f){
            for(i = 0; i < spatial; ++i){
                int index = b*filters*spatial + f*spatial + i;
                x[index] = (x[index] - mean[f])/(sqrt(variance[f]) + .000001f);
            }
        }
    }
}
 
void const_cpu(int N, float ALPHA, float *X, int INCX)
{
    int i;
    for(i = 0; i < N; ++i) X[i*INCX] = ALPHA;
}
 
void mul_cpu(int N, float *X, int INCX, float *Y, int INCY)
{
    int i;
    for(i = 0; i < N; ++i) Y[i*INCY] *= X[i*INCX];
}
 
void pow_cpu(int N, float ALPHA, float *X, int INCX, float *Y, int INCY)
{
    int i;
    for(i = 0; i < N; ++i) Y[i*INCY] = pow(X[i*INCX], ALPHA);
}
 
void axpy_cpu(int N, float ALPHA, float *X, int INCX, float *Y, int INCY)
{
    int i;
    for(i = 0; i < N; ++i) Y[i*INCY] += ALPHA*X[i*INCX];
}
 
void scal_cpu(int N, float ALPHA, float *X, int INCX)
{
    int i;
    for(i = 0; i < N; ++i) X[i*INCX] *= ALPHA;
}
 
void fill_cpu(int N, float ALPHA, float *X, int INCX)
{
    int i;
    for(i = 0; i < N; ++i) X[i*INCX] = ALPHA;
}
 
void deinter_cpu(int NX, float *X, int NY, float *Y, int B, float *OUT)
{
    int i, j;
    int index = 0;
    for(j = 0; j < B; ++j) {
        for(i = 0; i < NX; ++i){
            if(X) X[j*NX + i] += OUT[index];
            ++index;
        }
        for(i = 0; i < NY; ++i){
            if(Y) Y[j*NY + i] += OUT[index];
            ++index;
        }
    }
}
 
void inter_cpu(int NX, float *X, int NY, float *Y, int B, float *OUT)
{
    int i, j;
    int index = 0;
    for(j = 0; j < B; ++j) {
        for(i = 0; i < NX; ++i){
            OUT[index++] = X[j*NX + i];
        }
        for(i = 0; i < NY; ++i){
            OUT[index++] = Y[j*NY + i];
        }
    }
}
 
void copy_cpu(int N, float *X, int INCX, float *Y, int INCY)
{
    int i;
    for(i = 0; i < N; ++i) Y[i*INCY] = X[i*INCX];
}
 
void mult_add_into_cpu(int N, float *X, float *Y, float *Z)
{
    int i;
    for(i = 0; i < N; ++i) Z[i] += X[i]*Y[i];
}
 
void smooth_l1_cpu(int n, float *pred, float *truth, float *delta, float *error)
{
    int i;
    for(i = 0; i < n; ++i){
        float diff = truth[i] - pred[i];
        float abs_val = fabs(diff);
        if(abs_val < 1) {
            error[i] = diff * diff;
            delta[i] = diff;
        }
        else {
            error[i] = 2*abs_val - 1;
            delta[i] = (diff < 0) ? 1 : -1;
        }
    }
}
 
void l1_cpu(int n, float *pred, float *truth, float *delta, float *error)
{
    int i;
    for(i = 0; i < n; ++i){
        float diff = truth[i] - pred[i];
        error[i] = fabs(diff);
        delta[i] = diff > 0 ? 1 : -1;
    }
}
 
void softmax_x_ent_cpu(int n, float *pred, float *truth, float *delta, float *error)
{
    int i;
    for(i = 0; i < n; ++i){
        float t = truth[i];
        float p = pred[i];
        error[i] = (t) ? -log(p) : 0;
        delta[i] = t-p;
    }
}
 
void logistic_x_ent_cpu(int n, float *pred, float *truth, float *delta, float *error)
{
    int i;
    for(i = 0; i < n; ++i){
        float t = truth[i];
        float p = pred[i];
        error[i] = -t*log(p) - (1-t)*log(1-p);
        delta[i] = t-p;
    }
}
 
void l2_cpu(int n, float *pred, float *truth, float *delta, float *error)
{
    int i;
    for(i = 0; i < n; ++i){
        float diff = truth[i] - pred[i];
        error[i] = diff * diff;
        delta[i] = diff;
    }
}
 
float dot_cpu(int N, float *X, int INCX, float *Y, int INCY)
{
    int i;
    float dot = 0;
    for(i = 0; i < N; ++i) dot += X[i*INCX] * Y[i*INCY];
    return dot;
}
 
void softmax(float *input, int n, float temp, int stride, float *output)
{
    int i;
    float sum = 0;
    float largest = -FLT_MAX;
    for(i = 0; i < n; ++i){
        if(input[i*stride] > largest) largest = input[i*stride];
    }
    for(i = 0; i < n; ++i){
        float e = exp(input[i*stride]/temp - largest/temp);
        sum += e;
        output[i*stride] = e;
    }
    for(i = 0; i < n; ++i){
        output[i*stride] /= sum;
    }
}
 
 
void softmax_cpu(float *input, int n, int batch, int batch_offset, int groups, int group_offset, int stride, float temp, float *output)
{
    int g, b;
    for(b = 0; b < batch; ++b){
        for(g = 0; g < groups; ++g){
            softmax(input + b*batch_offset + g*group_offset, n, temp, stride, output + b*batch_offset + g*group_offset);
        }
    }
}
 
void upsample_cpu(float *in, int w, int h, int c, int batch, int stride, int forward, float scale, float *out)
{
    int i, j, k, b;
    for(b = 0; b < batch; ++b){
        for(k = 0; k < c; ++k){
            for(j = 0; j < h*stride; ++j){
                for(i = 0; i < w*stride; ++i){
                    int in_index = b*w*h*c + k*w*h + (j/stride)*w + i/stride;
                    int out_index = b*w*h*c*stride*stride + k*w*h*stride*stride + j*w*stride + i;
                    if(forward) out[out_index] = scale*in[in_index];
                    else in[in_index] += scale*out[out_index];
                }
            }
        }
    }
}

【图像处理】双线性插值法实现图像的缩放（C实现）

https://blog.csdn.net/weixin_43227685/article/details/88970457

#include <string.h>   
#include <math.h>     
#include <stdio.h>     
#include <stdlib.h>     
#include <malloc.h>  
#include<time.h>//时间相关头文件，可用其中函数计算图像处理速度  
 
#define   WIDTHBYTES(bits) (((bits)+31)/32*4)//用于使图像宽度所占字节数为4byte的倍数  
 
//#define MYDRAW_HEIGHT 986  //目标图像高度  
//define MYDRAW_WIDTH 1572  //目标图像宽度  
 
#define MYDRAW_HEIGHT 246  //目标图像高度
#define MYDRAW_WIDTH 393  //目标图像
 
typedef unsigned char  BYTE;
typedef unsigned short WORD;
typedef unsigned long  DWORD;
typedef long LONG;
 
 
 
//位图文件头信息结构定义  
//其中不包含文件类型信息（由于结构体的内存结构决定，要是加了的话将不能正确读取文件信息）  
 
typedef struct tagBITMAPFILEHEADER {
	DWORD  bfSize;          //文件大小  
	WORD   bfReserved1;     //保留字，不考虑  
	WORD   bfReserved2;     //保留字，同上  
	DWORD  bfOffBits;       //实际位图数据的偏移字节数，即前三个部分长度之和  
} BITMAPFILEHEADER;
 
//信息头BITMAPINFOHEADER，也是一个结构，其定义如下：  
 
typedef struct tagBITMAPINFOHEADER {
	//public:  
	DWORD   biSize;             //指定此结构体的长度，为40  
	LONG    biWidth;            //位图宽  
	LONG    biHeight;           //位图高  
	WORD    biPlanes;           //平面数，为1  
	WORD    biBitCount;         //采用颜色位数，可以是1，2，4，8，16，24，新的可以是32  
	DWORD   biCompression;      //压缩方式，可以是0，1，2，其中0表示不压缩  
	DWORD   biSizeImage;        //实际位图数据占用的字节数  
	LONG    biXPelsPerMeter;    //X方向分辨率  
	LONG    biYPelsPerMeter;    //Y方向分辨率  
	DWORD   biClrUsed;          //使用的颜色数，如果为0，则表示默认值(2^颜色位数)  
	DWORD   biClrImportant;     //重要颜色数，如果为0，则表示所有颜色都是重要的  
} BITMAPINFOHEADER;
 
void main()
{
	long now = 0;
	now = clock();//存储图像处理开始时间  
 
	BITMAPFILEHEADER bitHead, writebitHead;
	BITMAPINFOHEADER bitInfoHead, writebitInfoHead;
	FILE* pfile;//输入文件  
	FILE* wfile;//输出文件  
 
	char strFile[50] = "C:\\testpicture\\1.bmp";//打开图像路径，BMP图像必须为24位真彩色格式  
	char strFilesave[50] = "C:\\testpicture\\3.bmp";//处理后图像存储路径  
	fopen_s(&pfile, strFile, "rb");//文件打开图像  
	fopen_s(&wfile, strFilesave, "wb");//打开文件为存储修改后图像做准备  
									   //读取位图文件头信息  
	WORD fileType;
	fread(&fileType, 1, sizeof(WORD), pfile);
	fwrite(&fileType, 1, sizeof(WORD), wfile);
	if (fileType != 0x4d42)
	{
		printf("file is not .bmp file!");
		return;
	}
	//读取位图文件头信息  
	fread(&bitHead, 1, sizeof(tagBITMAPFILEHEADER), pfile);
	writebitHead = bitHead;//由于截取图像头和源文件头相似，所以先将源文件头数据赋予截取文件头  
						   //读取位图信息头信息  
    fread(&bitInfoHead, 1, sizeof(BITMAPINFOHEADER), pfile);
	writebitInfoHead = bitInfoHead;//同位图文件头相似  
 
	writebitInfoHead.biHeight = MYDRAW_HEIGHT;//为截取文件重写位图高度  
	writebitInfoHead.biWidth = MYDRAW_WIDTH;//为截取文件重写位图宽度  
	int mywritewidth = WIDTHBYTES(writebitInfoHead.biWidth*writebitInfoHead.biBitCount);//BMP图像实际位图数据区的宽度为4byte的倍数，在此计算实际数据区宽度  
	writebitInfoHead.biSizeImage = mywritewidth*writebitInfoHead.biHeight;//计算位图实际数据区大小  
 
	writebitHead.bfSize = 54 + writebitInfoHead.biSizeImage;//位图文件头大小为位图数据区大小加上54byte  
	fwrite(&writebitHead, 1, sizeof(tagBITMAPFILEHEADER), wfile);//写回位图文件头信息到输出文件  
	fwrite(&writebitInfoHead, 1, sizeof(BITMAPINFOHEADER), wfile);//写回位图信息头信息到输出文件  
 
	int width = bitInfoHead.biWidth;
	int height = bitInfoHead.biHeight;
	//分配内存空间把源图存入内存     
	int l_width = WIDTHBYTES(width*bitInfoHead.biBitCount);//计算位图的实际宽度并确保它为4byte的倍数  
	int write_width = WIDTHBYTES(writebitInfoHead.biWidth*writebitInfoHead.biBitCount);//计算写位图的实际宽度并确保它为4byte的倍数  
 
	BYTE    *pColorData = (BYTE *)malloc(height*l_width);//开辟内存空间存储图像数据  
	memset(pColorData, 0, height*l_width);
 
	BYTE    *pColorDataMid = (BYTE *)malloc(mywritewidth*MYDRAW_HEIGHT);//开辟内存空间存储图像处理之后数据  
	memset(pColorDataMid, 0, mywritewidth*MYDRAW_HEIGHT);
 
	long nData = height*l_width;
	long write_nData = mywritewidth*MYDRAW_HEIGHT;//截取的位图数据区长度定义  
 
												  //把位图数据信息读到数组里     
	fread(pColorData, 1, nData, pfile);//图像处理可通过操作这部分数据加以实现  
 
	/*******************图像处理部分******************/
	/*******************双线性插值******************/
	for (int hnum = 0; hnum < MYDRAW_HEIGHT; hnum++)
		for (int wnum = 0; wnum < MYDRAW_WIDTH; wnum++)
		{
			double d_original_img_hnum = hnum*height / (double)MYDRAW_HEIGHT;
			double d_original_img_wnum = wnum*width / (double)MYDRAW_WIDTH;
			int i_original_img_hnum = d_original_img_hnum;
			int i_original_img_wnum = d_original_img_wnum;
			double distance_to_a_x = d_original_img_wnum - i_original_img_wnum;//在原图像中与a点的水平距离  
			double distance_to_a_y = d_original_img_hnum - i_original_img_hnum;//在原图像中与a点的垂直距离  
 
			int original_point_a = i_original_img_hnum*l_width + i_original_img_wnum * 3;//数组位置偏移量，对应于图像的各像素点RGB的起点,相当于点A    
			int original_point_b = i_original_img_hnum*l_width + (i_original_img_wnum + 1) * 3;//数组位置偏移量，对应于图像的各像素点RGB的起点,相当于点B  
			int original_point_c = (i_original_img_hnum + 1)*l_width + i_original_img_wnum * 3;//数组位置偏移量，对应于图像的各像素点RGB的起点,相当于点C   
			int original_point_d = (i_original_img_hnum + 1)*l_width + (i_original_img_wnum + 1) * 3;//数组位置偏移量，对应于图像的各像素点RGB的起点,相当于点D   
			if (i_original_img_hnum +1== MYDRAW_HEIGHT - 1)
			{
				original_point_c = original_point_a;
				original_point_d = original_point_b;
			}
			if (i_original_img_wnum +1== MYDRAW_WIDTH - 1)
			{
				original_point_b = original_point_a;
				original_point_d = original_point_c;
			}
 
			int pixel_point = hnum*write_width + wnum * 3;//映射尺度变换图像数组位置偏移量  
			pColorDataMid[pixel_point] =
				pColorData[original_point_a] * (1 - distance_to_a_x)*(1 - distance_to_a_y) +
				pColorData[original_point_b] * distance_to_a_x*(1 - distance_to_a_y) +
				pColorData[original_point_c] * distance_to_a_y*(1 - distance_to_a_x) +
				pColorData[original_point_c] * distance_to_a_y*distance_to_a_x;
			pColorDataMid[pixel_point + 1] =
				pColorData[original_point_a + 1] * (1 - distance_to_a_x)*(1 - distance_to_a_y) +
				pColorData[original_point_b + 1] * distance_to_a_x*(1 - distance_to_a_y) +
				pColorData[original_point_c + 1] * distance_to_a_y*(1 - distance_to_a_x) +
				pColorData[original_point_c + 1] * distance_to_a_y*distance_to_a_x;
			pColorDataMid[pixel_point + 2] =
				pColorData[original_point_a + 2] * (1 - distance_to_a_x)*(1 - distance_to_a_y) +
				pColorData[original_point_b + 2] * distance_to_a_x*(1 - distance_to_a_y) +
				pColorData[original_point_c + 2] * distance_to_a_y*(1 - distance_to_a_x) +
				pColorData[original_point_c + 2] * distance_to_a_y*distance_to_a_x;
 
		}
	/*******************双线性插值******************/
	/*******************图像处理部分******************/
 
	fwrite(pColorDataMid, 1, write_nData, wfile);   //将处理完图像数据区写回文件  
	fclose(pfile);
	fclose(wfile);
 
	printf("图像处理完成\n");
	printf("运行时间为：%dms\n", int(((double)(clock() - now)) / CLOCKS_PER_SEC * 1000));//输出图像处理花费时间信息  
}

https://blog.csdn.net/py184473894/article/details/90739167

 

def bilinear_interpolation(img,scale):
    dst_cols=(int)(img.shape[0]*scale)
    dst_rows=(int)(img.shape[1]*scale)
    img_dst=np.zeros([dst_cols,dst_rows])
 
    for i in range(dst_cols-1):
        for j in range(dst_rows-1):
            #坐标转换
            scr_x=(i+0.5)/scale-0.5
            scr_y=(j+0.5)/scale-0.5
 
            #整数部分
            int_x=int(scr_x)
            #小数部分
            float_x=scr_x-int_x
 
            int_y=int(scr_y)
            float_y=scr_y-int_y
 
 
            if int_x==img.shape[0]-1:
                int_x_p=img.shape[0]-1
            else:
                int_x_p=int_x+1
 
            if int_y==img.shape[1]-1:
                int_y_p=img.shape[1]-1
            else:
                int_y_p=int_y+1
            
 
            img_dst[i][j]=(1-float_x)*(1-float_y)*img[int_x][int_y]+(1-float_x)*float_y*img[int_x][int_y_p]+\
                          float_x*(1-float_y)*img[int_x_p][int_y]+float_x*float_y*img[int_x_p][int_y_p]
 
    return img_dst