darknet_images Namespace Reference

Functions
	batch_detection (network, images, class_names, class_colors, thresh=0.25, hier_thresh=.5, nms=.45, batch_size=4)
	batch_detection_example ()
	check_arguments_errors (args)
	check_batch_shape (images, batch_size)
	convert2relative (image, bbox)
	image_classification (image, network, class_names)
	image_detection (image_or_path, network, class_names, class_colors, thresh)
	load_images (images_path)
	main ()
	parser ()
	perform_detection (args, network, class_names, class_colors)
	prepare_batch (images, network, channels=3)
	save_annotations (name, image, detections, class_names)

Function Documentation

batch_detection()

darknet_images.batch_detection	(		network,
			images,
			class_names,
			class_colors,
			thresh = 0.25,
			hier_thresh = .5,
			nms = .45,
			batch_size = 4
	)

Here is the call graph for this function:

Here is the caller graph for this function:

batch_detection_example()

darknet_images.batch_detection_example

(

)

Here is the call graph for this function:

check_arguments_errors()

darknet_images.check_arguments_errors

(

args

)

Here is the caller graph for this function:

check_batch_shape()

darknet_images.check_batch_shape	(		images,
			batch_size
	)

    Image sizes should be the same width and height

Here is the caller graph for this function:

convert2relative()

darknet_images.convert2relative	(		image,
			bbox
	)

YOLO format uses normalized coordinates for annotation.

Args:
    image: Input image (numpy array).
    bbox: Bounding box in absolute coordinates (x, y, width, height).

Returns:
    Tuple representing bounding box coordinates in relative format (x_rel, y_rel, w_rel, h_rel).

Here is the caller graph for this function:

image_classification()

darknet_images.image_classification	(		image,
			network,
			class_names
	)

Here is the call graph for this function:

image_detection()

darknet_images.image_detection	(		image_or_path,
			network,
			class_names,
			class_colors,
			thresh
	)

Here is the call graph for this function:

Here is the caller graph for this function:

load_images()

darknet_images.load_images

(

images_path

)

If image path is given, return it directly
For txt file, read it and return each line as image path
In other case, it's a folder, return a list with names of each
jpg, jpeg and png file

Here is the caller graph for this function:

main()

darknet_images.main

(

)

Here is the call graph for this function:

Here is the caller graph for this function:

parser()

darknet_images.parser

(

)

Here is the caller graph for this function:

perform_detection()

darknet_images.perform_detection	(		args,
			network,
			class_names,
			class_colors
	)

Here is the call graph for this function:

Here is the caller graph for this function:

prepare_batch()

darknet_images.prepare_batch	(		images,
			network,
			channels = 3
	)

Here is the call graph for this function:

Here is the caller graph for this function:

save_annotations()

darknet_images.save_annotations	(		name,
			image,
			detections,
			class_names
	)

Files saved with image_name.txt and relative coordinates.

Args:
    name: Name of the input image file.
    image: Input image (numpy array).
    detections: List of detected objects, each represented as (label, confidence, bbox).
    class_names: List of class names.

Saves:
    Text file with YOLO-style annotations for object detection.

Here is the call graph for this function:

Here is the caller graph for this function: