ahmetozlu / vehicle_counting_tensorflow

:oncoming_automobile: "MORE THAN VEHICLE COUNTING!" This project provides prediction for speed, color and size of the vehicles with TensorFlow Object Counting API.
https://www.youtube.com/watch?v=PrqnhHf6fhM
MIT License
888 stars 362 forks source link

How to get counting value of all the entered cars? #23

Open shivakutty opened 6 years ago

shivakutty commented 6 years ago

in the below example images,

  1. first white car detected 2.second not car not detected 3.again third car detected .. i want to count all the vehicle.

1 2 counter3

Akhtar303nu commented 6 years ago

Hi @shivakutty have you solved this problem

rakash commented 5 years ago

solution to this? @shivakutty

Hawkes1996 commented 5 years ago

How do I cancel the color prediction

rakash commented 5 years ago

@Hawkes1996 - cancel as in? one option is you can comment out the line.

Hawkes1996 commented 5 years ago

@ Hawkes1996 - 取消如何?一个选项是你可以注释掉这一行。

Comment out which line?

ChampionTej05 commented 5 years ago

The situation is so, because, As per the hardcoded logic written in the utils.speed_prediction.predict_speed function. When an object's current box's bottom location and it's previous box's bottom location is in the between 205 to 210 pixel range, it will be counted as detected. So if you just drag the line down, to 300 and check if the bottom lies between 290 to 310, you will get better results.