42 lines
1.0 KiB
Python
42 lines
1.0 KiB
Python
|
import base64
|
||
|
|
|
||
|
|
import cv2
|
||
|
|
import numpy as np
|
||
|
|
from flask import Flask, jsonify, request
|
||
|
|
|
||
|
|
from main import find_occupied_squares
|
||
|
|
from runner import find_homography, warp_board
|
||
|
|
from tensor_classifier import predict_board
|
||
|
|
|
||
|
|
app = Flask(__name__)
|
||
|
|
|
||
|
|
|
||
|
|
@app.route("/", methods=["POST"])
|
||
|
|
def process():
|
||
|
|
data = request.get_json(force=True)
|
||
|
|
|
||
|
|
decoded = base64.b64decode(data["img"])
|
||
|
|
img_array = np.frombuffer(decoded, dtype=np.uint8)
|
||
|
|
camera_img = cv2.imdecode(img_array, flags=cv2.COLOR_BGR2RGB)
|
||
|
|
camera_img = cv2.cvtColor(camera_img, cv2.COLOR_BGR2RGB)
|
||
|
|
|
||
|
|
# def do_everything:
|
||
|
|
homography = find_homography(camera_img)
|
||
|
|
warped_board = warp_board(camera_img, homography)
|
||
|
|
occupied_squares = find_occupied_squares(warped_board)
|
||
|
|
board = predict_board(occupied_squares)
|
||
|
|
|
||
|
|
# Finally, output for unity to read
|
||
|
|
return jsonify({
|
||
|
|
"homography": homography.tolist(),
|
||
|
|
"board": board.to_array,
|
||
|
|
})
|
||
|
|
|
||
|
|
|
||
|
|
def main():
|
||
|
|
app.run(host='0.0.0.0', debug=True)
|
||
|
|
|
||
|
|
|
||
|
|
if __name__ == '__main__':
|
||
|
|
main()
|