Image Histogram | Bright Image | Dark Image | Low Contrast Image | High Contrast Image

Image Histogram: In general, the histogram can be defined as the frequency of occurrence of data. In terms of image processing, the graphical representation of the digital image is called the histogram. It can also be defined as the graphical representation of pixel intensity values in a digital image.

The histogram helps in understanding the digital image. It analyses the digital image pixel by pixel (at the pixel level). Fig 1. displays the graphical representation (histogram) of the cameraman image.

The histogram has two axes i.e. X-axis and Y-axis respectively. The X-axis represents the total number of gray levels (L) ranging from 0 to L-1. The Y-axis represents the total number of pixels. In Fig 1., L=256 means it is a grayscale image i.e. X-axis ranges from 0 to 255.

The Matlab provides an in-built function i.e. "imhist" to draw the histogram of any digital image. Using below code one can draw the histogram of any digital image:
I = imread('cameraman.tif');
imhist(I)

Drawback: The histogram provides no information regarding the spatial distribution of an image’s pixel values.

Applications of Histograms

The digital image can be predicted based on its histogram.
The digital image can be analyzed based on its histogram.
It can be used in image thresholding.
It can be used in image enhancement.
It can be used to equalize an image.
The brightness and contrast of the digital image can be adjusted by its histogram.
If we have input and output histograms of an image, we can determine which type of transformation is applied in the algorithm.

Fig 1. Cameraman image with the histogram

Histogram of Bright Image

The components of the histogram are biased towards the high side of the grayscale as shown in Fig 2. This shows that the brightness is high in the current image (Fig 2).

Fig 2. Bright cameraman image with the histogram

Histogram of Dark Image

The components of the histogram are concentrated on the low (dark) side of the grayscale as shown in Fig 3. This shows that the brightness is very low (dark image) in the current image (Fig 3).

Fig 3. Dark cameraman image with the histogram

Histogram of Low Contrast Image

The histogram will be narrow and will be centered towards the middle of the grayscale as shown in Fig 3. This shows that the contrast is low in the current image (Fig 4).

Fig 4. Low contrast cameraman image with the histogram

Histogram of High Contrast Image

The components of the histogram cover a broad range of the grayscale as shown in Fig 4. This shows that the contrast is high in the current image (Fig 5).