sachadee

import os
import cv2
import numpy as np
from util import detectionclassOV as net

detection = net.Detection("./weights/compiled_detection.blob")

def main():
    frame = cv2.imread('./images/plate.jpg')
    image = frame.copy()

sachadee

import os
import cv2
import numpy as np
from util import detectionclass as net

detection = net.Detection('./weights/detection.onnx')

def main():
    frame = cv2.imread('./images/plate.jpg')
    image = frame.copy()

sachadee

import math
import os
import cv2
import numpy as np 
from pyclipper import *
from shapely.geometry import Polygon


class Detection:
    def __init__(self, onnx_path, session=None):

sachadee

inflect
librosa==0.9.2
matplotlib
numpy
Pillow
PyQt5
scikit-learn
scipy
sounddevice
SoundFile==0.10.3.post1

sachadee

from Crypto.Cipher import AES
from Crypto.Random import get_random_bytes
import base64

def encrypt_aes_gcm(plaintext, key):
    cipher = AES.new(key, AES.MODE_GCM)
    ciphertext, tag = cipher.encrypt_and_digest(plaintext)
    return ciphertext, cipher.nonce, tag

# Example usage

sachadee

//Function to get to convert to bytes the base64 values from python

function base64ToUint8Array(base64) {
		var binaryString = atob(base64);
		var len = binaryString.length;
		var bytes = new Uint8Array(len);
		for (var i = 0; i < len; i++) {
			bytes[i] = binaryString.charCodeAt(i);
		}
		return bytes;

sachadee

 with torch.no_grad():
        for idx, (image, _) in enumerate(
            tqdm(loader, desc="Create embeddings matrix", total=len(loader)),
        ):
            embeddings = np.empty([1,512])
            embeddings[int(0) :] = F.normalize(backbone(image.to(device))).cpu()
            image = image[0].permute(1,2,0)
            imgarr = image.cpu().detach().numpy()
            print(imgarr.dtype)
            opencvImage = cv2.cvtColor(imgarr, cv2.COLOR_RGB2BGR)

sachadee

# get embedding for each face
    embeddings = np.zeros([len(loader.dataset), embedding_size])
    with torch.no_grad():
        for idx, (image, _) in enumerate(
            tqdm(loader, desc="Create embeddings matrix", total=len(loader)),
        ):
            print(idx)
            embeddings[idx, :] = F.normalize(backbone(image.to(device))).cpu()
            tensor_to_pil = transforms.ToPILImage()(image.squeeze_(0)).convert('RGB')
            tensor_to_pil.show()

sachadee

tensor_to_pil = transforms.ToPILImage()(tensor)
tensor_to_pil.show()

sachadee

#Base Image
print("Image Dtype:", np.array(image).dtype)
print("Image shape:", np.array(image).shape)

print('***********')

#Tensor image
print("Tensor Dtype:", tensor.dtype)
print("Tensor shape:", tensor.shape)