10种软件滤波方法的示例程序

10种软件滤波方法的示例程序(JKRL)文章编号:541----加入日期:2003-8-18
      『关闭窗口』
      10种软件滤波方法的示例程序(JKRL)

        假定从8位AD中读取数据(如果是更高位的AD可定义数据类型为int),子程序为get_ad();

        1、限副滤波
        /*  A值可根据实际情况调整
            value为有效值,new_value为当前采样值  
            滤波程序返回有效的实际值  */
        #define A 10

        char value;

        char filter()
        {
           char  new_value;
           new_value = get_ad();
           if ( ( new_value - value > A ) || ( value - new_value > A )
              return value;
           return new_value;
                
        }

        2、中位值滤波法
        /*  N值可根据实际情况调整
            排序采用冒泡法*/
        #define N  11

        char filter()
        {
           char value_buf[N];
           char count,i,j,temp;
           for ( count=0;count           {
              value_buf[count] = get_ad();
              delay();
           }
           for (j=0;j           {
              for (i=0;i              {
                 if ( value_buf[i]>value_buf[i+1] )
                 {
                    temp = value_buf[i];
                    value_buf[i] = value_buf[i+1];
                     value_buf[i+1] = temp;
                 }
              }
           }
           return value_buf[(N-1)/2];
        }    

        3、算术平均滤波法
        /*
        */

        #define N 12

        char filter()
        {
           int  sum = 0;
           for ( count=0;count           {
              sum + = get_ad();
              delay();
           }
           return (char)(sum/N);
        }

        4、递推平均滤波法(又称滑动平均滤波法)
        /*
        */
        #define N 12

        char value_buf[N];
        char i=0;

        char filter()
        {
           char count;
           int  sum=0;
           value_buf[i++] = get_ad();
           if ( i == N )   i = 0;
           for ( count=0;count              sum = value_buf[count];
           return (char)(sum/N);
        }

        5、中位值平均滤波法(又称防脉冲干扰平均滤波法)
        /*
        */
        #define N 12

        char filter()
        {
           char count,i,j;
           char value_buf[N];
           int  sum=0;
           for  (count=0;count           {
              value_buf[count] = get_ad();
              delay();
           }
           for (j=0;j           {
              for (i=0;i              {
                 if ( value_buf[i]>value_buf[i+1] )
                 {
                    temp = value_buf[i];
                    value_buf[i] = value_buf[i+1];
                     value_buf[i+1] = temp;
                 }
              }
           }
           for(count=1;count              sum += value[count];
           return (char)(sum/(N-2));
        }

        6、限幅平均滤波法
        /*
        */  
        略 参考子程序1、3

        7、一阶滞后滤波法
        /* 为加快程序处理速度假定基数为100,a=0~100 */

        #define a 50

        char value;

        char filter()
        {
           char  new_value;
           new_value = get_ad();
           return (100-a)*value + a*new_value;
        }

        8、加权递推平均滤波法
        /* coe数组为加权系数表,存在程序存储区.*/

        #define N 12

        char code coe[N] = {1,2,3,4,5,6,7,8,9,10,11,12};
        char code sum_coe = 1+2+3+4+5+6+7+8+9+10+11+12;

        char filter()
        {
           char count;
           char value_buf[N];
           int  sum=0;
           for (count=0,count           {
              value_buf[count] = get_ad();
              delay();
           }
           for (count=0,count              sum += value_buf[count]*coe[count];
           return (char)(sum/sum_coe);
        }

        9、消抖滤波法

        #define N 12

        char filter()
        {
           char count=0;
           char new_value;
           new_value = get_ad();
           while (value !=new_value);
           {
              count++;
              if (count>=N)   return new_value;
               delay();
              new_value = get_ad();
           }
           return value;    
        }

        10、限幅消抖滤波法
        /*
        */
        略 参考子程序1、9