CameraObstacleDetector Class Reference

Simple obstacle detection algorithm for grayscale camera images. More...

#include <CameraObstacleDetector.h>

Inheritance diagram for CameraObstacleDetector:

Collaboration diagram for CameraObstacleDetector:

Classes
struct	Result

Public Member Functions
	CameraObstacleDetector (uint8_t threshold=60, float trigger=0.15f)
Result	process (const uint8_t *image, size_t width, size_t height)
Public Member Functions inherited from MessageSource
void	subscribe (MessageHandler &handler, MessageOrigin origin=MessageOrigin::Undefined, MessageContent content=MessageContent::Undefined)
	Subscribe a message handler to this source, with optional filtering.
void	unsubscribeAll ()
	Remove all registered message handlers.
void	sendMessage (Message< float > &msg)
	Publish a message to all registered handlers.
void	sendMessage (const Message< Coordinate< float > > &msg)
	Publish a message to all registered handlers.
void	sendMessage (const Message< GPSCoordinate > &msg)
	Publish a message to all registered handlers.
void	sendMessage (const Message< MotionState3D > &msg)
	Overload for MotionState3D messages.

Protected Attributes
uint8_t	_threshold
float	_trigger
Protected Attributes inherited from MessageSource
std::vector< MessageHandlerEntry >	messageHandlers_

Detailed Description

Simple obstacle detection algorithm for grayscale camera images.

This class scans a central region of a grayscale image (e.g., from a camera or vision sensor) to detect the presence of dark obstacles based on pixel intensity and density. It is useful for basic obstacle avoidance in robotics.

Features:

• Scans the center region of the image for dark pixels
• Triggers detection if the density of dark pixels exceeds a threshold
• Returns detection status and density (fraction of dark pixels)
• Configurable pixel threshold and trigger density
• Lightweight, suitable for embedded/Arduino use

Note

Expected format:

• The input image should be a grayscale buffer (uint8_t*)
• The pixel intensity is expected to be in the range 0 (black) to 255 (white).
• The layout of the image buffer should be row-major (i.e., pixel at (x,y) is at index y*width + x).
• on ESP32 e.g. PIXFORMAT_GRAYSCALE, FRAMESIZE_QVGA (320 x 240)

Usage Example:

CameraObstacleDetector detector(60, 0.2f); // threshold=60, trigger=20%
auto result = detector.process(image, width, height);
if (result.detected) {
  // Obstacle detected, take action
}

---

The documentation for this class was generated from the following file:

• src/TinyRobotics/sensors/CameraObstacleDetector.h