gives an image consisting of the first n rows of image.


gives an image consisting of the last n rows of image.


gives rows row1 through row2.


gives the image that spans row1 to row2 and col1 to col2.


returns a video in which every frame consists of the specified region of interest.


gives the 3D image consisting of the specified slices, rows, and columns.


  • ImageTake is typically used for extracting regions of interest by cropping or trimming an image.
  • ImageTake uses the standard sequence specification:
  • Allall elements
    nelements 1 up to n
    UpTo[n]elements 1 up to n
    -nlast n elements
    {n}element n only
    {m,n}elements m through n inclusive
    {m,n,s}elements m through n in steps of s
  • If fewer elements exist than are requested, ImageTake will just return the elements that are present.
  • For video inputs, the region of interest can be a fixed specification for all frames or a function of time (in seconds from the start).


open allclose all

Basic Examples  (2)

Take the first 70 pixel rows from an image:

Take rows 30 through 100 and columns 60 through 130 from an image:

Scope  (5)

Data  (3)

Take rows 80 to 200 from a grayscale image:

Take the top 100 rows of every video frame:

Take the top 23 slices of a 3D image:

Parameters  (2)

Take the last 100 rows from an image:

Take rows 30 to 100:

Take row 100:

Take every other row of the image:

Take the first 100 columns:

Take a range of rows and columns:

Take the bottom 23 slices of a 3D image:

Take a selected horizontal slice of a 3D image:

Take all slices in a selected region of the 3D image:

Take selected part of a 3D volume:

Applications  (2)

Trim an image by removing a 30-pixel-wide border:

Use ImageKeypoints to determine a region of interest:

Highlight the region of interest:

Convert from image coordinates to matrix coordinates and extract the region of interest:

Properties & Relations  (2)

Use ImageTake to obtain the same top-bottom reflection as with ImageReflect:

Use Downsample or ImageTake to downsample an image by extracting every other pixel:

Interactive Examples  (2)

Interactive image panning:

Interactively take specified parts on a volume:

Wolfram Research (2008), ImageTake, Wolfram Language function, (updated 2021).


Wolfram Research (2008), ImageTake, Wolfram Language function, (updated 2021).


Wolfram Language. 2008. "ImageTake." Wolfram Language & System Documentation Center. Wolfram Research. Last Modified 2021.


Wolfram Language. (2008). ImageTake. Wolfram Language & System Documentation Center. Retrieved from


@misc{reference.wolfram_2022_imagetake, author="Wolfram Research", title="{ImageTake}", year="2021", howpublished="\url{}", note=[Accessed: 09-December-2022 ]}


@online{reference.wolfram_2022_imagetake, organization={Wolfram Research}, title={ImageTake}, year={2021}, url={}, note=[Accessed: 09-December-2022 ]}