Camera

Functions
int	camera_open ()
int	camera_open_black ()
int	camera_open_at_res (enum Resolution res)
int	camera_open_device (int number, enum Resolution res)
int	camera_open_device_model_at_res (int number, enum Model model, enum Resolution res)
int	camera_load_config (const char *name)
int	get_camera_width (void)
int	get_camera_height (void)
int	camera_update (void)
pixel	get_camera_pixel (point2 p)
int	get_channel_count (void)
int	get_object_count (int channel)
const char *	get_object_data (int channel, int object)
int	get_object_data_length (int channel, int object)
double	get_object_confidence (int channel, int object)
int	get_object_area (int channel, int object)
rectangle	get_object_bbox (int channel, int object)
point2	get_object_centroid (int channel, int object)
point2	get_object_center (int channel, int object)
void	camera_close ()
void	set_camera_config_base_path (const char *const path)
const unsigned char *	get_camera_frame_row (unsigned row)
const unsigned char *	get_camera_frame ()

Detailed Description

The functions here deal with the camera. Specifically, the functions here deal with object detection through the use of color.

Setup

To use the object detections, you'll first have to have a .conf file. On a wombat, you can go to settings, then channels, then add a configuration. Next, edit that configuration and add a new channel (for the color object detection, it should be HSV Blob Tracking). Finally, configure that channel, selecting the color that you want the camera to detect when detecting objects on that channel.

Example object detection

Assuming you have a configuration file "botball.conf" and you have a Channel 0 on that file configured to detect red pom poms, then the following code would print 'red' when a red pom is seen.

#include <kipr/wombat.h>
#include <stdio.h>
 
int main(){
     camera_load_config("botball.conf");
     camera_open();
     
     // wait a little and update the camera
     int i = 0;
     for (i = 0; i < 10; i++){
         camera_update();
         msleep(10);
     }
 
     int red_object_seen = 0;
     while (!red_object_seen){
         // get the bounding box of the largest object in
         // channel 0 (the red channel)
         rectangle object_bounding_box = get_object_bbox(0, 0);
         if (object_bounding_box.width * object_bounding_box.height != 0){
             printf("red object seen!");
             red_object_seen = 1;
         }
 
         // update the camera
         camera_update();
         msleep(10);
     }
 
     camera_close();  // cleanup the camera
     return 0;
}

Function Documentation

camera_close()

void camera_close

(

)

Cleanup the current camera instance.

See also

camera_open

camera_open_at_res

camera_open_device

camera_load_config()

int camera_load_config

(

const char *

name

)

Loads the config file specified by name. The system will look for the config in the base path.

Parameters

name	The configuration to load. Configuration file names are case sensitive.

Note

Do NOT include the config file extension ".conf" in the name parameter.

Returns

1 on success, 0 on failure.

See also

set_camera_config_base_path

camera_open()

int camera_open

(

)

Opens the default system camera for use at LOW_RES (160x120).

Returns

1 on success, 0 on failure

See also

camera_open_at_res

camera_open_device

camera_close

camera_open_at_res()

int camera_open_at_res

(

enum Resolution

res

)

Opens the default system camera for use at a given resolution.

Parameters

res	The resolution the camera should operate at. This can be: LOW_RES (160x120) MED_RES (320x240) HIGH_RES (640x480) TELLO_RES (1280x720)

Warning

Only LOW_RES is currently supported. The function will fail for other resolutions.

Returns

1 on success, 0 on failure

See also

camera_open

camera_open_device

camera_close

camera_open_black()

int camera_open_black

(

)

Opens the default system camera for use at LOW_RES (160x120). This will improve frame rates for the black Logitech camera

Returns

1 on success, 0 on failure

See also

camera_open_at_res

camera_open_device

camera_close

camera_open_device()

int camera_open_device	(	int	number,
		enum Resolution	res
	)

Opens a camera for use.

Parameters

number	The camera's id. 0 is the first camera, 1 is the second camera, etc.
res	The resolution the camera should operate at. This can be: LOW_RES (160x120) MED_RES (320x240) HIGH_RES (640x480) TELLO_RES (1280x720)

Warning

Only LOW_RES is currently supported. The function will fail for other resolutions.

Returns

1 on success, 0 on failure

See also

camera_open

camera_close

camera_open_device_model_at_res()

int camera_open_device_model_at_res	(	int	number,
		enum Model	model,
		enum Resolution	res
	)

Opens a camera for use.

Parameters

number	The camera's id. 0 is the first camera, 1 is the second camera, etc.
res	The resolution the camera should operate at. This can be: LOW_RES (160x120) MED_RES (320x240) HIGH_RES (640x480) TELLO_RES (1280x720)
model	The camera model WHITE_2016 The white 2016 Botball camera BLACK_2017 The black 2017 Botball camera TELLO Tello drone camera

Warning

MED_RES is supported only for the BLACK_2017 camera and HIGH_RES is not supported

Returns

1 on success, 0 on failure

See also

camera_open

camera_close

camera_update()

int camera_update

(

void

)

Pulls a new image from the camera for processing.

Returns

1 on success, 0 on failure.

get_camera_frame()

const unsigned char* get_camera_frame

(

)

Retrieves the current camera frame as a BGR (BGR888) array. The returned pointer is invalid after camera_update() is called again.

Returns

the current BGR888 camera frame.

get_camera_frame_row()

const unsigned char* get_camera_frame_row

(

unsigned

row

)

Retrieves the current camera frame row as a BGR (BGR888) array. The returned pointer is invalid after camera_update() is called again.

Returns

the current BGR888 camera frame row.

get_camera_height()

int get_camera_height

(

void

)

Gets the camera's y resolution.

Attention

This value might be different than the previously set y resolution. Never assume the y resolution.

Returns

The camera's y resolution, in pixels.

get_camera_pixel()

pixel get_camera_pixel

(

point2

p

)

Gets the color of a pixel.

Parameters

p	The point at which the pixel lies.

Returns

The rgb value of the pixel located at point p.

Note

A (r, g, b) value of (-1, -1, -1) will be returned for points that are out of range.

get_camera_width()

int get_camera_width

(

void

)

Gets the camera's x resolution.

Attention

This value might be different than the previously set x resolution. Never assume the x resolution.

Returns

The camera's x resolution, in pixels.

get_channel_count()

int get_channel_count

(

void

)

Returns

Number of channels in the current configuration.

See also

get_object_count

get_object_area()

int get_object_area	(	int	channel,
		int	object
	)

Returns

The object's bounding box area. -1 is returned if the channel or object doesn't exist.

get_object_bbox()

rectangle get_object_bbox	(	int	channel,
		int	object
	)

Returns

The bounding box of the given object on the given channel.

get_object_center()

point2 get_object_center	(	int	channel,
		int	object
	)

Returns

The (x, y) center of the given object on the given channel.

get_object_centroid()

point2 get_object_centroid	(	int	channel,
		int	object
	)

Returns

The (x, y) centroid of the given object on the given channel.

get_object_confidence()

double get_object_confidence	(	int	channel,
		int	object
	)

Returns

The confidence, between 0.0 and 1.0, that given object on the given channel is significant. If the channel or object doesn't exist, 0.0 is returned.

get_object_count()

int get_object_count

(

int

channel

)

Parameters

channel

The channel to scan for objects.

Note

Objects are sorted by area, largest first.

Returns

Number of objects in the given channel, -1 if channel doesn't exist.

See also

get_channel_count

get_object_data()

const char* get_object_data	(	int	channel,
		int	object
	)

Returns

The string data associated with a given object on a given channel. If there is no data associated, 0 is returned.

Note

This data is not guaranteed to be null terminated.

This string pointer will be invalid after a call to camera_update()

See also

get_object_data_length

get_object_data_length()

int get_object_data_length	(	int	channel,
		int	object
	)

Returns

The length of the string data associated with a given object on a given channel. If there is no data associated, 0 is returned. If the channel or object is invalid, 0 is returned.

See also

get_object_data

set_camera_config_base_path()

void set_camera_config_base_path

(

const char *const

path

)

Sets the path in which to look for camera configurations.

Parameters

path	the absolute directory path in which to look for camera configurations.