代码很简单，很容易看懂。
import pandas as pdimport matplotlib.pyplot as plt
#headers = ['Name', 'Age', 'Marks']df = pd.read_csv('samples-folder/samples2.csv')
df=df.drop(0)df=df.drop(1)# print(df)
if(len(df.columns) == 3): df.columns = ['sample interval', 'signal', 'filtered'] # print(df)
list1 = df.filtered list2 = df.signal# print(list1)
else: df.columns = ['sample interval', 'signal'] list1 = df.signal
plotting1 = []plotting2 = []for i in list1: plotting1.append(float(i))
# for i in list2:# plotting2.append(float(i))
# print(type(plotting[0]))
# print(plotting)# print(len(plotting))
length_df = len(df)
x = []for i in range(1,length_df+1): x.append(i) plt.figure(figsize=(300,70))
# plt.plot(x, plotting2, label ="Original ECG signal") # plt.show()
plt.figure(figsize=(300,70))plt.plot(x[:], plotting1[:], label ="Filtered ECG signal")plt.show()
# plt.figure(figsize=(400,50))# plt.subplot(1,2,1)# plt.plot(x, plotting1)# plt.subplot(1,2,2)# plt.plot(x, plotting2)
# df.set_index('Marks').plot()
# plt.show()haar = []
count = 28flag = 1
i = 1while(i<= length_df): j = i # if(i == 10000-count+1):# for k in range(1, count):# haar.append(-1)# break while(i<= length_df and i < j + count): haar.append(1) i+=1 while(i<= length_df and i< j + (2*count)): haar.append(-1) i+=1 flag = - flag # print(haar)# print(len(haar))plt.figure(figsize=(200,50))plt.plot(x, plotting1, label ="ecg signal")plt.plot(x, haar, label ="haar")plt.show()
list_peaks = []i = 0while(i< length_df): sum1 = 0 j = i while(i< length_df and i < j+ count): sum1 += plotting1[i] i+=1 while(i< length_df and i< j + (2* count)): sum1 -= plotting1[i] i+=1 list_peaks.append(sum1) sum1 = 0
# print(list_peaks)
# print(len(list_peaks))base = []for i in range(0, len(list_peaks)): base.append(i+1) plt.figure(figsize=(10,6))plt.plot(base, list_peaks)
list_peaks_stuffed = []
index = 0
# def compute():# global index# list_peaks_stuffed = []# for i in range(len(list_peaks)-1):# j = i# while(i< j+ count):# list_peaks_stuffed.append(0)# i+=1# if(index < len(list_peaks)):# list_peaks_stuffed.append(list_peaks[index])# index +=1
# while(i< j+ (2* count) - 1):# list_peaks_stuffed.append(0)# i+=1
# i = j# return list_peaks_stuffed[0]
# list_new = []# for k in range(0, 10000):# val = compute() # left = 10000 - len(list_peaks_stuffed) -1
# for i in range(0, left//2):# list_peaks_stuffed.append(0)
# list_peaks_stuffed.append(list_peaks[len(list_peaks) -1])
# for j in range(0, left//2):# list_peaks_stuffed.append(0) # to_be_popped = len(list_peaks_stuffed) - 10000# for i in range(to_be_popped):# list_peaks_stuffed.pop()
# # print(len(list_peaks_stuffed))# print(list_peaks_stuffed)# # plt.subplot(1,2,2)# plt.figure(figsize=(10,6))# plt.plot(x, list_peaks_stuffed)# plt.show()
import math
length = len(plotting1) -1left_len = length
pointer = 0remaining_len = length+1
x = []for i in range(length+1): x.append(i+1)
iter_count = 1arr = []fig, ax = plt.subplots(1, 1, figsize=(60, 20))# ax[0].plot(x, plotting1)# ax[0].set_title("Original")while(iter_count<=1 ): arr = arr[left_len+1:] arr3 = [] arr_d = [] for k in range(0, length+1): arr_d.append(0) for k in range(0, length+1): arr3.append(0) arr2 = [] pointer1 = pointer count = 0 while(pointer1 < left_len ): arr2.append((plotting1[pointer1] + plotting1[pointer1 + 1])/ math.sqrt(2)) pointer1 +=2 count += 1 pointer2 = pointer while(pointer2 < left_len): arr2.append((plotting1[pointer2] - plotting1[pointer2+ 1] )/ math.sqrt(2)) pointer2 += 2 if((left_len + 1)%2 == 1): arr2.append(plotting1[length]) # if(iter_count >= 1 and iter_count <= 10): for g in range(0, len(arr2)): if(2*g <= length): arr3[2*g] = arr2[g] for g in range(count, len(arr2)): if(2*(g - count) <= length): arr_d[2*(g- count)] = arr2[g] for g in range(0, length+1): if(arr3[g] == 0): if((g-1) <= length and (g+1) <= length): arr3[g] = (arr3[g-1] + arr3[g+1])/2 for g in range(0, length+1): if(arr_d[g] == 0): if((g-1) <= length and (g+1) <= length): arr_d[g] = (arr_d[g-1] + arr_d[g+1])/2 arr = arr2 + arr ax.plot(x, arr3) ax.plot(x, arr_d) ax.plot(x, plotting1) # ax[iter_count -1].set_title(f"Iteration{iter_count}")# print("Hello world")
iter_count +=1 left_len = ((left_len + 1)//2)-1 remaining_len = length - left_len
import math
length = len(plotting1) -1left_len = length
pointer = 0remaining_len = length+1
x = []for i in range(length+1): x.append(i+1)
iter_count = 1arr = []fig_, ax_ = plt.subplots(1, 1, figsize=(30, 5))# ax[0].plot(x, plotting1)
arr = arr[left_len+1:]arr3 = []arr_d = []for k in range(0, length+1): arr_d.append(0)
for k in range(0, length+1): arr3.append(0)
arr2 = []pointer1 = pointercount = 0while(pointer1 < left_len ): arr2.append((plotting1[pointer1] + plotting1[pointer1 + 1])/ math.sqrt(2)) pointer1 +=2 count += 1
pointer2 = pointerwhile(pointer2 < left_len): arr2.append((plotting1[pointer2] - plotting1[pointer2+ 1] )/ math.sqrt(2)) pointer2 += 2
if((left_len + 1)%2 == 1): arr2.append(plotting1[length])
# if(iter_count >= 1 and iter_count <= 10):
print(len(arr2))print(len(arr))for g in range(0, len(arr2)): if(2*g <= length): arr3[2*g] = arr2[g]
for g in range(count, len(arr2)): if(2*(g - count) <= length): arr_d[2*(g- count)] = arr2[g]
for g in range(0, length+1): if(arr3[g] == 0): if((g-1) <= length and (g+1) <= length): arr3[g] = (arr3[g-1] + arr3[g+1])/2
for g in range(0, length+1): if(arr_d[g] == 0): if((g-1) <= length and (g+1) <= length): arr_d[g] = (arr_d[g-1] + arr_d[g+1])/2
arr = arr2 + arr
ax_.plot(x, arr3)ax_.plot(x, arr_d)ax_.plot(x, plotting1)
ax_.set_title("Combined plotting of signals")
# print(len(arr_d), len(plotting1))
arr_t_wave = []for samp in range(0, length_df): if(samp+20 < length_df-1): if(arr_d[samp + 20]- arr_d[samp] >= 0.015): arr_t_wave.append(0) elif(plotting1[samp+20] - arr_d[samp] < -0.015): arr_t_wave.append(0) else: arr_t_wave.append((plotting1[samp])) else: arr_t_wave.append(plotting1[samp]) if(arr_t_wave[samp] <= 0): arr_t_wave[samp] = 0 else: arr_t_wave[samp] = (arr_t_wave[samp]) * 0.4
#Removing R peaks from arr_t_wave:arr_t_wave2 = []for samp in range(0, length_df): if(samp+10 < length_df-1): if(arr_d[samp + 10]- arr_d[samp] >= 0.015): arr_t_wave2.append(arr_d[samp]* 3) elif(plotting1[samp+10] - arr_d[samp] < -0.015): arr_t_wave2.append(arr_d[samp]* 3) else: arr_t_wave2.append((plotting1[samp])) else: arr_t_wave2.append(plotting1[samp]) if(arr_t_wave2[samp] <= 0): arr_t_wave2[samp] = 0 else: arr_t_wave2[samp] = (arr_t_wave2[samp]) * 0.4 # if( samp+ 10 < 9989 and (arr_d[samp] >= 0.15 or arr_d[samp + 10] >= 0.15)):# samp1 = samp -1000# while samp1 <= samp:# arr_t_wave[samp1] = 0# samp1 += 1# while samp1 <= samp+ 5:# arr_t_wave.append(0)# samp1 += 1 # samp +=5
# else:# arr_t_wave[samp] = 0
ax_.plot(arr_t_wave2)dict1 = {}
marker_list = []
s = 0limit = 200while(s <= length_df): t = s+1 flag = 0 while(t< length_df and t <= s + limit and arr_t_wave2[t] <= 0.008 and arr_t_wave2[t] >= -0.008): flag += 1 t += 1 if(flag == limit): flag = 0 st = s s = s+ limit while(s< length_df and arr_t_wave2[s] <= 0.008 and arr_t_wave2[s] >= -0.008): s += 1 dict1[st] = s marker_list.append(st) marker_list.append(s-30) else: s = s + 1
for i in marker_list: if marker_list.count(i + 30) >= 1: marker_list = list(filter(lambda x: x != i and x != i+ 30, marker_list))
marker_list_f = []
p = 0while(p < len(marker_list)-1): val = (marker_list[p] + marker_list[p+1])//2 marker_list_f.append(val) p += 2
marker_signal = []
for i in range(0, length_df): if i in marker_list: marker_signal.append(1) else: marker_signal.append(0) marker_signal_f = []
for i in range(0, length_df): if i in marker_list_f: marker_signal_f.append(1) else: marker_signal_f.append(0) # print(marker_list)# print(marker_list_f)# print(dict1)fig1, ax1 = plt.subplots(1, 1, figsize=(30, 5))
# ax1[iter_count -1].plot(x, plotting1)ax1.plot(x, plotting1)ax1.plot(x, arr_d)
ax1.set_title("R peak detector signal")
fig4, ax4 = plt.subplots(1, 1, figsize=(20, 5))
# ax1[iter_count -1].plot(x, plotting1)ax4.plot(x, plotting1)ax4.plot(x, arr_t_wave)ax4.set_title("Prominent peaks detector signal")
fig3, ax3 = plt.subplots(1, 1, figsize=(20, 5))
# ax1[iter_count -1].plot(x, plotting1)ax3.plot(x, plotting1)ax3.plot(x, arr_t_wave2)ax3.set_title("Combined detector signals")
# fig5, ax5 = plt.subplots(1, 1, figsize=(20, 5))
# # ax1[iter_count -1].plot(x, plotting1)# ax5.plot(x[1200: 2500], plotting1[1200: 2500])# ax5.plot(x[1200: 2500], arr_t_wave2[1200: 2500])# ax5.plot(x[1200: 2500], marker_signal[1200: 2500])
fig6, ax6 = plt.subplots(1, 1, figsize=(20, 5))
# ax1[iter_count -1].plot(x, plotting1)ax6.plot(x, plotting1)ax6.plot(x, marker_signal_f)ax6.set_title("Segmented signal")
# fig2, ax2 = plt.subplots(1, 1, figsize=(60, 15))# x1 = []# plotting2 = []# arr_d1 = []# for i in range(700, 1200):# x1.append(i+1)# plotting2.append(plotting1[i])# arr_d1.append(arr_d[i])
# arr_t_wave = []# for samp in range(700, 1200):# if(samp+10 < 1200):# if(arr_d[samp + 15]- arr_d[samp] >= 0.015):# arr_t_wave.append(0)# elif(plotting1[samp+15] - arr_d[samp] < -0.015):# arr_t_wave.append(0) # else:# arr_t_wave.append((plotting1[samp]))# else:# arr_t_wave.append(plotting1[samp]) # if(arr_t_wave[samp-700] <= 0):# arr_t_wave[samp-700] = 0# else:# arr_t_wave[samp-700] = (arr_t_wave[samp-700]* arr_t_wave[samp-700]) # # ax1[iter_count -1].plot(x, plotting1)# ax2.plot(x1, plotting2)# ax2.plot(x1, arr_d1)
# fig4, ax4 = plt.subplots(1, 1, figsize=(60, 15))# ax4.plot(x1, plotting2)# ax4.plot(x1, arr_t_wave)
marker_list_f.insert(0, 0)marker_list_f.append(length_df -1)
print(marker_list_f)i = 0fig8, ax8 = plt.subplots(len(marker_list_f) -1, 1, figsize=(3, 50))
while(i < len(marker_list_f)-1): beat = [] for j in range(marker_list_f[i], marker_list_f[i+1] + 1): beat.append(plotting1[j]) x = [] for k in range(len(beat)): x.append(k + 1) ax8[i].plot(x, beat) ax8[i].set_title(f"Beat{i+1}") i += 1
工学博士，担任《Mechanical System and Signal Processing》《中国电机工程学报》《控制与决策》等期刊审稿专家，擅长领域：现代信号处理，机器学习，深度学习，数字孪生，时间序列分析，设备缺陷检测、设备异常检测、设备智能故障诊断与健康管理PHM等。