分段读取csv文件并可视化处理

Posted 2022-12-07 空中旋转篮球

1.数据

使用数据为csv文件，数据有89万条记录，名称字段具体如下：

Time (sec)	Delta Time (sec)	Segment ID	Latitude (deg)	Longitude (deg)	Easting (m)	Northing (m)	Cross-Track (m)	Along-Track (m)	Height (m HAE)	Height (m MSL)	Classification	Signal Confidence	Dem_height	luccID
32.56000081	78976682.22	161266	29.08135819	113.4059348	734190.7155	3219390.276	-22.7432411	20201.67926	77.74754333	93.76404762	0	0	0	0
32.56000081	78976682.22	161266	29.08135858	113.4059346	734190.7011	3219390.318	-22.7526257	20201.72316	91.41132355	107.4278278	0	0	0	0
32.56000081	78976682.22	161266	29.08135802	113.4059348	734190.722	3219390.257	-22.7389931	20201.65946	71.5628891	87.57939339	0	0	0	0
32.56000081	78976682.22	161266	29.08135773	113.4059349	734190.733	3219390.224	-22.7318915	20201.62621	61.22312164	77.23962593	0	0	0	0
32.56000081	78976682.22	161266	29.08135811	113.4059348	734190.7185	3219390.267	-22.7412967	20201.67021	74.91683197	90.93333626	0	0	0	0
32.56000081	78976682.22	161266	29.08135616	113.4059355	734190.7915	3219390.052	-22.6938061	20201.44813	5.772859573	21.78936386	0	0	0	0
32.56000081	78976682.22	161266	29.08135614	113.4059355	734190.7922	3219390.05	-22.6933262	20201.44587	5.074115753	21.09062004	0	0	0	0
32.56000081	78976682.22	161266	29.08135651	113.4059354	734190.7784	3219390.091	-22.7023598	20201.48811	18.22643089	34.24293518	0	0	0	0
32.56000081	78976682.22	161266	29.08135656	113.4059354	734190.7765	3219390.096	-22.7035772	20201.49374	19.99853134	36.01503563	0	0	0	0
32.56000081	78976682.22	161266	29.08135584	113.4059356	734190.8036	3219390.016	-22.6859499	20201.41132	-5.66560459	10.3508997	0	0	0	0
32.56010081	78976682.22	161266	29.08136468	113.405934	734190.6294	3219390.994	-22.7452324	20202.4024	80.65776825	96.67429392	0	0	0	0

2.读取数据 

2.1完整读取数据并显示

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt

inputpath=r"E:\\csv_dbscan\\ATL03_20200703015729_01180802_005_01_gt1r.csv"
df=pd.read_csv(inputpath)
X=df['Time (sec)']
Y=df['Height (m HAE)']

plt.figure()
plt.scatter(X,Y,marker='o',s=0.000003,label='Point Cloud')
plt.legend()
plt.show()

放大显示：

 2.2分段读取数据并显示

（1）根据总记录数和分段数来计算每段的数量是多少，然后我们每次只读取一段数据。

方法1：使用chunksize参数实现分段读取和显示

chunksize是按照行记录数量来分段的

df_chunker = pd.read_csv(inputpath,chunksize=segment_count)
for df_item in df_chunker:
    X_seg = df_item['Time (sec)']
    Y_seg = df_item['Height (m HAE)']
    plt.figure()
    plt.scatter(X_seg, Y_seg, marker='o', s=1, label='Point Cloud')
    plt.legend()
    plt.show()

比如我们每段数据设置为segment_count=10000条，那么就有segment_n=N/segment_count(segment_count每段记录数，N总记录数，segment_n为段数）段数据。

我们按照顺序读取每段数据并显示：

方法2：使用沿线距离来分段显示 

比如我们使用Along-Track (m)属性1000m分段显示：第一段的范围是20201.67926-20301.67926，后面依次加1000.

len_data=len(df['Time (sec)']) #行记录数
segment_count=1000
Along_track_n=int((df['AlongTrack'][len_data-1]-df['AlongTrack'][0])/segment_count)+1
start=df['AlongTrack'][0]
end=df['AlongTrack'][len_data-1]
for len_seg in range(Along_track_n):
    df_seg = df.loc[(df['AlongTrack'] >= (start + len_seg * segment_count)) & (df['AlongTrack'] <= (start+ (len_seg + 1) * segment_count)),:]
    X_seg = df_seg['AlongTrack']
    Y_seg = df_seg['Height (m HAE)']
    if len(X_seg)==0:
        print("该段没数据！")
    plt.figure()
    plt.scatter(X_seg, Y_seg, marker='o', s=1, label='Point Cloud')
    plt.legend()
    plt.show()