解析SCUT FIR Pedestrian Dataset数据

一 、 前言

最近解析SCUT FIR Pedestrian Dataset数据，发现只有matlab的解析代码，使用起来非常不方便，开源的教程基本没有Python的解析，这里自己写了一个，

二 、解析代码

2.1 解析官方给的txt文档

# 传入官方txt文档路径，返回解析好的字典格式内容

def process_txt(txtpath):
    """
    传入原始txt文件路径，解析文件

	walk_person ： 走路的人
	ride_person ：骑车的人
	squat_person ：蹲着的人
	people： 一群人
	person? ：不确定是否是人
	people? ： 不确定是否是一群人
	
	
	目标存坐标存储在 pos 标签中，
	4个值的含义分别是 左上角坐标x,y  目标w,h

    """
    with open(txtpath,'r') as f:
        info = f.readlines()

    stap = 7
    label_dict = {}
    N = (len(info)-3)//7
    
    for i in range(N):
        info_ = info[3+stap*i : 3+ stap*(i+1)]
        occl = info_[5][6:-3].split(" ")
        lock = info_[6][6:-3].split(" ")
        posv = info_[4][6:-3].split(";")
        pos = info_[3][6:-3].split(";")

        label =info_[2].split(" ")[0][5:-1]
        strr =int(info_[2].split(" ")[1][4:])
        end =int(info_[2].split(" ")[2][4:])

        for idx ,frameid in enumerate(range(strr,end+1)):
            label_ ={}
            label_['frame'] = str(frameid)
            label_['label'] = str(label)
            label_['pos'] = pos[idx]
            label_['posv'] = posv[idx]
            label_['lock'] = lock[idx]
            label_['occl'] = occl[idx]

            if str(frameid) not in list(label_dict.keys()):
                label_dict[str(frameid)] = [label_]
            else:
                label_dict[str(frameid)].append(label_)

    return label_dict

2.1 拆解视频，成为独立的帧

在这里插入代码片
import os
import json
import fnmatch
import shutil



def open_save(file, savepath):
    # read .seq file, and save the images into the savepath
    f = open(file, 'rb')
    string = f.read().decode('latin-1')
    splitstring = "\xFF\xD8\xFF\xE0\x00\x10\x4A\x46\x49\x46"
    # split .seq file into segment with the image prefix
    strlist = string.split(splitstring)
    f.close()
    count = 0
    # delete the image folder path if it exists
    if os.path.exists(savepath):
        shutil.rmtree(savepath)
    # create the image folder path
    if not os.path.exists(savepath):
        os.mkdir(savepath)
    # deal with file segment, every segment is an image except the first one
    for img in strlist:
        filename = str(count) + '.jpg'
        filenamewithpath = os.path.join(savepath, filename)
        # abandon the first one, which is filled with .seq header
        if count > 0:
            i = open(filenamewithpath, 'wb+')
            i.write(splitstring.encode('latin-1'))
            i.write(img.encode('latin-1'))
            i.close()
        count += 1

# 解析视频成为jpg图片
def Split_Video_to_Image():
    rootdir = "./SCUT FIR Pedestrian Dataset/videos/set02"
    # rootdir = "E:\\GPassport\\Work\\Search\\PedestrianDetection\\CaltechDatasets\\set\\set01"
    # walk in the rootdir, take down the .seq filename and filepath
    for parent, dirnames, filenames in os.walk(rootdir):
        for filename in filenames:
            # check .seq file with suffix
            if fnmatch.fnmatch(filename, '*.seq'):
                # take down the filename with path of .seq file
                thefilename = os.path.join(parent, filename)
                # create the image folder by combining .seq file path with .seq filename
                thesavepath = parent + '/' + filename.split('.')[0]
                print("Filename=" + thefilename)
                print("Savepath=" + thesavepath)
                open_save(thefilename, thesavepath)

2.3 批量生成txt对应的json标签文件

def all_txt_process():
    txtroot = './SCUT FIR Pedestrian Dataset/annotations'
    savejson = './SCUT FIR Pedestrian Dataset/annotations_label'

    for file in os.listdir(txtroot):
        for txtname in os.listdir(os.path.join(txtroot,file)):
            if os.path.splitext(txtname)[-1] != '.txt':
                continue
            label_dict = process_txt(os.path.join(txtroot,file,txtname))

            svaepath = os.path.join(savejson,file)
            os.makedirs(svaepath,exist_ok=True)

            with open(os.path.join(svaepath,txtname.replace('.txt','_frame_label.json')), 'w') as f:
                json.dump(label_dict, f,)

2.4解析格式