Namespace containing functions to assist in object detection. More...

Classes
struct	Object
	Represents a single detected object. Combines attributes from TorchObject and TensorflowObject structs. More...

Enumerations
enum class	ObjectDetectionMethod { eUnknown , eTorch }
	Enum class to define the different object detection methods available. More...

enum class	ObjectDetectionType { eUnknown , eMallet , eWaterBottle }
	Enum class to define the different object detection types available. More...

Functions
cv::Point2f	FindObjectCenter (const Object &stObject)
	Find the center of an object.

void	EstimatePoseFromCameraFrame (Object &stTag)
	Estimate the pose of a tag from a camera frame.

Detailed Description

Namespace containing functions to assist in object detection.

Author: clayjay3 (clayt.nosp@m.onra.nosp@m.ycowe.nosp@m.n@gm.nosp@m.ail.c.nosp@m.om)

Date: 2025-05-05

Enumeration Type Documentation

◆ ObjectDetectionMethod

enum class objectdetectutils::ObjectDetectionMethod

strong

Enum class to define the different object detection methods available.

Author: clayjay3 (clayt.nosp@m.onra.nosp@m.ycowe.nosp@m.n@gm.nosp@m.ail.c.nosp@m.om)

Date: 2025-05-05

    {
        eUnknown,    // Unknown detection method.
        eTorch,      // Torch detection using a YOLO model.
    };

◆ ObjectDetectionType

enum class objectdetectutils::ObjectDetectionType

strong

Enum class to define the different object detection types available.

Author: clayjay3 (clayt.nosp@m.onra.nosp@m.ycowe.nosp@m.n@gm.nosp@m.ail.c.nosp@m.om)

Date: 2025-05-13

    {
        eUnknown,       // Unknown detection type.
        eMallet,        // Detection is a mallet.
        eWaterBottle    // Detection is a water bottle.
    };

Function Documentation

◆ FindObjectCenter()

cv::Point2f objectdetectutils::FindObjectCenter ( const Object & stObject )

inline

Find the center of an object.

Parameters

stObject - The object to find the center of.

Returns: cv::Point2f - The center of the object.

Author: clayjay3 (clayt.nosp@m.onra.nosp@m.ycowe.nosp@m.n@gm.nosp@m.ail.c.nosp@m.om)

Date: 2025-05-05

    {
        // Find the center of the object.
        return cv::Point2f(stObject.pBoundingBox->x + stObject.pBoundingBox->width / 2, stObject.pBoundingBox->y + stObject.pBoundingBox->height / 2);
    }

◆ EstimatePoseFromCameraFrame()

void objectdetectutils::EstimatePoseFromCameraFrame ( Object & stTag )

inline

Estimate the pose of a tag from a camera frame.

Parameters

stTag - The tag to estimate the pose of.

Note: In order for this to be accurate, the camera's horizontal field of view (HFOV) and the camera frame size must be known.

Author: sam_hajdukiewicz (saman.nosp@m.thah.nosp@m.ajduk.nosp@m.iewi.nosp@m.cz@gm.nosp@m.ail..nosp@m.com) :3

Date: 2025-04-04

    {
        // Use camera field of view and camera frame size to determine tag angle in degrees from center of camera.
        double dDegreesPerPixel = stTag.dHorizontalFOV / stTag.cvImageResolution.width;
        // Find tag error in pixels from center of image.
        double dTagErrorX = (stTag.pBoundingBox->x + stTag.pBoundingBox->width / 2) - (stTag.cvImageResolution.width / 2);
        // Find angle error.
        double dTagAngleX = dTagErrorX * dDegreesPerPixel;
        // Reassign yaw and distance to tag.
        stTag.dYawAngle = dTagAngleX;
 
        // For the distance, we'll just use the screen percentage of the tag.
        stTag.dStraightLineDistance = (stTag.pBoundingBox->area() / (stTag.cvImageResolution.width * stTag.cvImageResolution.height)) * 100.0;
    }

Here is the caller graph for this function:

Classes

Enumerations