Image Statistics Functions#
The image statistics functions assist with understanding the dataset. It can be used to get a big picture view of the dataset and it’s underlying distribution.
dimensionstats#
What is the dimensionstats function#
The dimensionstats function is an aggregate metric that calculates various dimension based statistics for each individual image:
width
height
channels
size
aspect_ratio
depth
This function can be used in conjunction with the Outliers class to determine if there are any issues with any of the images in the dataset.
hashstats#
What is the hashstats function#
The hashstats function is an aggregate metric that calculates various hash values for each individual image:
This function can be used in conjunction with the Duplicates class in order to identify duplicate images.
labelstats#
What is the labelstats function#
The labelstats function provides summary statistics across classes and labels:
label_counts_per_class
label_counts_per_image
image_counts_per_label
image_indices_per_label
image_count
label_count
class_count
pixelstats#
What is the pixelstats function#
The pixelstats function is an aggregate metric that calculates normal statistics about pixel values for each individual image:
mean
std
var
skew
kurtosis
entropy
percentiles
histogram
This function can be used in conjunction with the Outliers class to determine if there are any issues with any of the images in the dataset.
visualstats#
What is the visualstats function#
The visualstats function is an aggregate metric that calculates visual quality statistics for each individual image:
brightness
sharpness
contrast
darkness
missing (as a percentage of total pixels)
zeros (as a percentage of total pixels)