stef_buwalda/ProjectIOT
1
0
Fork 0
mirror of https://github.com/StefBuwalda/ProjectIOT.git synced 2025-10-29 10:49:58 +00:00
Code Issues Packages Projects Releases Wiki Activity

Test script

Browse Source
This commit is contained in:
Stef
2025-06-19 20:32:43 +02:00
parent 96859fd2bb
commit 850a9f2b37
2 changed files with 36 additions and 9 deletions
Download Patch File Download Diff File Expand all files Collapse all files

16
OCR YOLO11/yolomain.py
View File

@@ -2,7 +2,10 @@ from ultralytics import YOLO
from PIL import Image
from transformers import TrOCRProcessor, VisionEncoderDecoderModel
import os
os.chdir("C:/Users/celma/OneDrive - Hanze/School/periode 1.4/IOT/YOLO11/License Plate Recognition.v11i.yolov11")
os.chdir(
"C:/Users/celma/OneDrive - Hanze/School/periode 1.4/IOT/YOLO11/License Plate Recognition.v11i.yolov11"
)
model = YOLO("license_plate_detector.pt")
@@ -17,18 +20,13 @@ clss = results[0].boxes.cls.cpu().tolist()
print(boxes)
if boxes is not None:
for box, cls in zip(boxes,clss):
crop_obj = boxes[int(box[1]):int(box[3]) + int(box[0]):int(box[2])]
for box, cls in zip(boxes, clss):
crop_obj = boxes[int(box[1]) : int(box[3]) + int(box[0]) : int(box[2])]
cropped_img = img.crop(
(int(box[0]), int(box[1]), int(box[2]), int(box[3]))
)
cropped_img = img.crop((int(box[0]), int(box[1]), int(box[2]), int(box[3])))
cropped_img.show()
processor = TrOCRProcessor.from_pretrained("microsoft/trocr-base-handwritten")
model = VisionEncoderDecoderModel.from_pretrained("microsoft/trocr-large-printed")
image = cropped_img.convert("RGB")
pixel_values = processor(images=image, return_tensors="pt").pixel_values
generated_ids = model.generate(pixel_values)

29
test2.py Normal file
View File

@@ -0,0 +1,29 @@
from ultralytics import YOLO
from PIL import Image
from transformers import TrOCRProcessor, VisionEncoderDecoderModel
import os
from datetime import datetime
import torch
print("11")
processor = TrOCRProcessor.from_pretrained(
"microsoft/trocr-base-handwritten", use_fast=True
)
print("12")
model = VisionEncoderDecoderModel.from_pretrained(
"microsoft/trocr-Large-printed"
)
print("13")
cropped_img = Image.open("license_plate.jpg")
cropped_img.show()
image = cropped_img.convert("RGB")
first = datetime.now()
pixel_values = processor(images=image, return_tensors="pt").pixel_values
generated_ids = model.generate(pixel_values)
print(generated_ids)
text = processor.batch_decode(generated_ids, skip_special_tokens=True)[0]
print(datetime.now() - first)
print(text)