请教特征提取的一些函数,有些地方不太明白

V2EX = way to explore

V2EX 是一个关于分享和探索的地方

已注册用户请登录

这是一个创建于 752 天前的主题，其中的信息可能已经有所发展或是发生改变。

import numpy as np


# 均方根
def featureRMS(data):
    return np.sqrt(np.mean(data ** 2, axis=0))


# 取绝对值求各列均值
def featureMAV(data):
    return np.mean(np.abs(data), axis=0)

# 求离散的差值，按列的后尾元素向上面的元素进行求差，然后取绝对值再求所有矩阵同一元素位置的和，最后除以矩阵个数
def featureWL(data):
    return np.sum(np.abs(np.diff(data, axis=0)), axis=0) / data.shape[0]

#
def featureZC(data, threshold=10e-7):
    numOfZC = []
    channel = data.shape[1]
    length = data.shape[0]

    for i in range(channel):
        count = 0
        for j in range(1, length):
            diff = data[j, i] - data[j - 1, i]
            mult = data[j, i] * data[j - 1, i]

            if np.abs(diff) > threshold and mult < 0:
                count = count + 1
        numOfZC.append(count / length)
    return np.array(numOfZC)


def featureSSC(data, threshold=10e-7):
    numOfSSC = []
    channel = data.shape[1]
    length = data.shape[0]

    for i in range(channel):
        count = 0
        for j in range(2, length):
            diff1 = data[j, i] - data[j - 1, i]
            diff2 = data[j - 1, i] - data[j - 2, i]
            sign = diff1 * diff2

            if sign > 0:
                if (np.abs(diff1) > threshold or np.abs(diff2) > threshold):
                    count = count + 1
        numOfSSC.append(count / length)
    return np.array(numOfSSC)

能请教一下 MAV 、WL 、ZC 、SSC4 个函数的都是啥数学公式？网上搜的那些都搜不到😅

大学数学早就还给老师了.....

4 条回复