Creating JSON object from Python's dictionary

Question

The following code was used to create JSON. But I have \\ in JSON.

"[{\"1_245_183_424_225_251_188_416_219_SJA4549J.jpeg\": {\"fileref\": \"\", \"size\": 0, \"filename\": \"1_245_183_424_225_251_188_416_219_SJA4549J.jpeg\", \"file_attributes\": {}, \"regions\": {\"plate\": {\"shape_attributes\": {\"name\": \"polygon\", \"all_points_x\": [], \"all_points_y\": []}, \"region_attributes\": {\"width\": 720, \"height\": 720, \"depth\": 3}}}}}, {\"1_245_183_424_225_251_188_416_219_SJA4549J.jpeg\": {\"fileref\": \"\", \"size\": 0, \"filename\": \"1_245_183_424_225_251_188_416_219_SJA4549J.jpeg\", \"file_attributes\": {}, \"regions\": {\"textline\": {\"shape_attributes\": {\"name\": \"polygon\", \"all_points_x\": [], \"all_points_y\": []}, \"region_attributes\": {\"width\": 720, \"height\": 720, \"depth\": 3}}}}}, {\"IMG_20200214_081450.jpeg\": {\"fileref\": \"\", \"size\": 0, \"filename\": \"IMG_20200214_081450.jpeg\", \"file_attributes\": {}, \"regions\": {\"plate\": {\"shape_attributes\": {\"name\": \"polygon\", \"all_points_x\": [], \"all_points_y\": []}, \"region_attributes\": {\"width\": 720, \"height\": 720, \"depth\": 3}}}}}, {\"IMG_20200214_081450.jpeg\": {\"fileref\": \"\", \"size\": 0, \"filename\": \"IMG_20200214_081450.jpeg\", \"file_attributes\": {}, \"regions\": {\"plate\": {\"shape_attributes\": {\"name\": \"polygon\", \"all_points_x\": [], \"all_points_y\": []}, \"region_attributes\": {\"width\": 720, \"height\": 720, \"depth\": 3}}}}}, {\"IMG_20200214_081450.jpeg\": {\"fileref\": \"\", \"size\": 0, \"filename\": \"IMG_20200214_081450.jpeg\", \"file_attributes\": {}, \"regions\": {\"textline\": {\"shape_attributes\": {\"name\": \"polygon\", \"all_points_x\": [], \"all_points_y\": []}, \"region_attributes\": {\"width\": 720, \"height\": 720, \"depth\": 3}}}}}, {\"IMG_20200214_081450.jpeg\": {\"fileref\": \"\", \"size\": 0, \"filename\": \"IMG_20200214_081450.jpeg\", \"file_attributes\": {}, \"regions\": {\"textline\": {\"shape_attributes\": {\"name\": \"polygon\", \"all_points_x\": [], \"all_points_y\": []}, \"region_attributes\": {\"width\": 720, \"height\": 720, \"depth\": 3}}}}}]"

Then I also have "[ at the beginning and ]" at end of file in printing to JSON file. Why I have those at the beginning and end of JSON file and how to remove?

The code is as follow.

import os
import xml.etree.ElementTree as ET
import pickle
from os import listdir, getcwd
from os.path import join
import numpy as np
import cv2
import math
import random
import json
xmlspath="/home/itc/Data/NumberPlate/PlateDetection/annotations/xmls/"


xmlFiles = [f for f in listdir(xmlspath) if f.endswith("." + "xml")];
num_objs = 0
iou = 0.0
num_hits = 0
train = []
val = []
for idx,xmlFile in enumerate(xmlFiles):
   if(idx > 10):
      break
   tree = ET.parse(xmlspath+xmlFile)
   root = tree.getroot()
   filename = xmlFile.split(".")[0]+".jpeg"
   size = root.find("size")
   width = int(size.find("width").text)
   height = int(size.find("height").text)
   depth = int(size.find("depth").text)
   for obj in root.iter("object"):
      newitem={
  "name":{"fileref":"","size":0,"filename":"","file_attributes":{},"regions":{"attribute":{"shape_attributes":{"name":"polygon","all_points_x":[],"all_points_y":[]},"region_attributes":{"width":0,"height":0,"depth":3}}}}  
}
      newitem[filename]=newitem.pop("name")
      newitem[filename]["filename"]=filename
      bndbox = obj.find("bndbox")
      category = obj.find("name")
      xmin=int(bndbox.find("xmin").text)
      ymin=int(bndbox.find("ymin").text)
      xmax=int(bndbox.find("xmax").text)
      ymax=int(bndbox.find("ymax").text)
      if(obj.find("name").text == "plate"):
         newitem[filename]["regions"]["plate"] = newitem[filename]["regions"].pop("attribute")
         newitem[filename]["regions"]["plate"]["region_attributes"]["width"]=width
         newitem[filename]["regions"]["plate"]["region_attributes"]["height"]=height  
      elif(obj.find("name").text == "textline"):
         newitem[filename]["regions"]["textline"] = newitem[filename]["regions"].pop("attribute")
         newitem[filename]["regions"]["textline"]["region_attributes"]["width"]=width
         newitem[filename]["regions"]["textline"]["region_attributes"]["height"]=height                  
      if(idx%8 == 0):
         val.append(newitem)
      else:
         train.append(newitem)
      print(idx)
      

jsontrain = json.dumps(train)
jsonval = json.dumps(val)
with open("numplate/train/train.json", "w") as outfile:
    json.dump(jsontrain, outfile)
with open("numplate/val/val.json", "w") as outfile:
    json.dump(jsonval, outfile)
      
      

Answer 1

The reason that you have "[ at the beginning and ]" at end of file" is that because your objects (val an train) are lists (not dictionaries)

First you have to convert them:

train = {"items" : train}
val =  = {"items" : val}

Then, for your encoding issue you can force the json.dump() function to not add the "" symbol by setting the ensure_ascii to False :

with open("numplate/train/train.json", 'w', encoding='utf8') as outfile:
    json.dump(train, outfile, ensure_ascii=False)