Final project of AI class. A platform who can recognize the dishes and show the heat.
A single convolutional network simultaneously predicts multiple bounding boxes and class probabilities for those boxes. YOLO trains on full images and directly optimizes detection performance.
-
YOLO is extremely fast. We simply run our neural network on a new image at test time to predict detections.
-
YOLO reasons globally about the image when making predictions.
-
YOLO learns generalizable representations of objects.
npm run test
<script src="https://cdnjs.cloudflare.com/ajax/libs/p5.js/1.0.0/p5.min.js"></script>
<script src="https://cdnjs.cloudflare.com/ajax/libs/p5.js/0.9.0/addons/p5.dom.min.js"></script>
<script src="https://unpkg.com/[email protected]/dist/ml5.min.js"></script>
<script src="https://code.jquery.com/jquery-3.4.1.min.js"></script>let fruit = new Array(
"apple",
"banana",
"cherry",
"grape",
"kiwi",
"orange",
"peach",
"pear",
"plum",
"pineapple",
"strawberry",
"broccoli",
"carrot",
"cucumber",
"cabbage",
"lettuce",
"tomato",
"potato"
);
let calo = new Array(
52,
87,
5,
3,
2,
45,
42,
96,
30,
96,
38,
45,
30,
14,
26,
10,
25,
110
);function imageReady() {
yolo.detect(upimg, gotResult);
}{
"label": "apple",
"confidence": 0.9158242344856262,
"h": 0.24185683177067682,
"w": 0.24617767333984375,
"x": 0.7303961240328275,
"y": 0.5583128195542556
}let name = new Array(); //Build an array of all detected labels
let num = new Array(); //Add counters for each labels
let q = 0;
for (let i = 0; i < objects.length; i++) {
//The first one is always special
if (i == 0) {
name[q] = objects[i].label;
num[q] = 1;
}
//Comparing the label with previous one
//A new label
else if (i > 0 && objects[i].label != objects[i - 1].label) {
q++;
name[q] = objects[i].label;
num[q] = 1;
}
//An existing label
else if (i > 0 && objects[i].label == objects[i - 1].label) {
num[q]++;
}
}for (h = 0; h <= fruit.length; h++) {
if (name[i] == fruit[h]) {
sum = sum + calo[h] * num[i].toFixed(1); //The total heat
document.getElementById("heat").innerHTML =
document.getElementById("heat").innerHTML +
calo[h].toFixed(1) +
" × " +
num[i] +
" cal<br>";
}
}
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent