Original subtitles

In this section, we will explain how to use the rule of seven to detect object classes, images, videos

and webcams will be used for detection.

However, before we begin the section, we will explain some arguments that can be used.

First there are argument weights.

The argument is the use of seven weights file used.

YOLO v seven is the default value.

For an example of its application.

See this?

The following argument is can't stress this argument is an object confidence threshold.

The default value is 0.25.

An example of its use is as follows.

The detection results that will be displayed on those with a score greater than this threshold.

Image size is the next argument.

This is the argument for inference size.

The default setting is 640.

This is an example of its application.

By increasing the size, it may increase the accuracy, but the process is slower.

Next is the source argument.

This argument that accepts the source to be detected.

This source could be either a file or a folder, An image, video or webcam can also be used as this

source.

The default value is inference images folder.

This is an example of its application.

Next argument fuel emits.

This is an argument for displaying the detection resolve.

There is no default value.

Simply enter the arguments to use it.

The following argument is safety XD.

This is an argument for saving the detection results to a file.

There is no default value.

Simply enter the arguments to use it.

The next video was so an example of the detection.

See you in the next video.