ValueError: X.shape[1] = 400 should be equal to 5994, the number of features at training time
what to do it has to do with svc/base.py file?

HI
I run ALPR.python
I received the value erroor, what should I do? Thank you

I have an issue with localization package that you used in line:
label_image = measure.label(localization.binary_car_image)
then i get this Error:

`> Traceback (most recent call last):

File "medium2.py", line 31, in
label_image = measure.label(localization.binary_car_image)
AttributeError: module 'localization' has no attribute 'binary_car_image'`

what's your solution?
thanks

Hi there,

I've install all of your requirements, but still doesn't run the tool. do you know why?

Hello @femioladeji. Thanks for creating and sharing this useful application. Can you please share your thoughts on integrating this application with a Python-based multimedia database system like EVA? Do you anticipate any use-cases for a query language over the extracted labels?

https://github.com/georgia-tech-db/license-plate-recognition/blob/main/README.ipynb
https://github.com/georgia-tech-db/eva

CREATE UDF IF NOT EXISTS OCRExtractor
INPUT (frame NDARRAY UINT8(3, ANYDIM, ANYDIM))
OUTPUT (labels NDARRAY STR(ANYDIM), bboxes NDARRAY FLOAT32(ANYDIM, 4), scores NDARRAY FLOAT32(ANYDIM))
TYPE TextExtraction
IMPL 'ocr_extractor.py';

SELECT OCRExtractor(data) FROM MyImages WHERE ocrextractor.label LIKE "LEM446A4"

@femioladeji : I have done installation based on your README file and everything is nicely installed. Now the moement I try to run ALPR.py file it is not allowing me to access any of the functionality of GUI. I am trying to give image file as input using GUI , however unable to do so.

**Attached is the screen shot for your information. Would be really happy if you can help me out with this. **

Why it doesn't work for motorcycles number plate detection

from skimage import measure
from skimage.measure import regionprops
import matplotlib.pyplot as plt
import matplotlib.patches as patches
import localization

this gets all the connected regions and groups them together

label_image = measure.label(localization.binary_car_image)

fig, (ax1) = plt.subplots(1)
ax1.imshow(localization.gray_car_image, cmap="gray");

regionprops creates a list of properties of all the labelled regions

for region in regionprops(label_image):
if region.area < 50:
#if the region is so small then it's likely not a license plate
continue

# the bounding box coordinates
minRow, minCol, maxRow, maxCol = region.bbox
rectBorder = patches.Rectangle((minCol, minRow), maxCol-minCol, maxRow-minRow, edgecolor="red", linewidth=2, fill=False)
ax1.add_patch(rectBorder)
# let's draw a red rectangle over those regions

plt.show()

AttributeError Traceback (most recent call last)
in ()
6
7 # this gets all the connected regions and groups them together
----> 8 label_image = measure.label(localization.binary_car_image)
9
10 fig, (ax1) = plt.subplots(1)

AttributeError: module 'localization' has no attribute 'binary_car_image'

Everything else installed fine until I ran python ALPR.py and received the fatal
ImportError: No module named 'Tkinter'

This seems to be an endemic issue in Python. Tried spinning up various pyenvs including 2.7.10 which is supposed to have a stable version of TKinter. Here's all the versions I tried :

2.7.0
2.7.10
2.7.17

No avail on any of these. Any insights on which 2.7.x works?

I was thinking of doing something like this a while back..
Why not add support for getting the plate number owner details and for verification.

Nice work bdw.

Heyy! I am getting this error while running the code. It is displaying everytime that number plate is not localised. Could you please help me out!!

While executing

pip install -r requirements.txt

I am getting following error -
Collecting functools32==3.2.3.post2 (from -r requirements.txt (line 11))
Using cached https://files.pythonhosted.org/packages/c5/60/6ac26ad05857c601308d8fb9e87fa36d0ebf889423f47c3502ef034365db/functools32-3.2.3-2.tar.gz
Complete output from command python setup.py egg_info:
This backport is for Python 2.7 only.

----------------------------------------

Command "python setup.py egg_info" failed with error code 1 in C:\Users\Dheeraj\AppData\Local\Temp\pip-install-81aa7vit\functools32\

NameError Traceback (most recent call last)
in ()
29 print("Cross Validation Result for ", str(num_of_fold), " -fold")
30 print(accuracy_result * 100)
---> 31 current_dir = os.path.dirname(os.path.realpath(file))
32 training_dataset_dir = os.path.join(current_dir, 'train')
33 image_data, target_data = read_training_data(training_dataset_dir)

NameError: name 'file' is not defined

I ran the command python ALPR.py and the gui window appears but there it says no image to display . In the top left it says, File Path : No Image File Chosen .
The test image output should be shown i guess why it is not visible , any leads on it .

Any help would be great
Thanks .

Could you provide the dataset with all the source vehicle images?

When I execute the project via cmd, the GUI shows up but nothing is responding i.e. its not reading the test image. I just can't interact with the GUI in any way.

When try to test your code with your test image, I get below error:

File "D:\carplate\License-Plate-Recognition-Nigerian-vehicles-master\deepMachine.py", line 33, in classify_objects
result_index = self.letters.index(result[0])
ValueError: b'E' is not in list

I'm currently looking in your blog. I have done installing the dependencies. I'm now with the localization.py step and downloading the car.jpg and in the directory of /home/nigel/license-plate-recognition, where should I create the localization.py?

Do I have to clone it first the full project?

Please help me, thanks!

Good day Mr. Femi

Please is it possible to run this code without the use of gui? I am talking of something like when I run the full.py, program, there is already a path for the image and it automatically shows me the license plate text already extracted instead of running the gui and pressing 'RUN ALPR' all the time

I'll be expecting a response.