Repeating Patterns as Spatial Frequencies
The scan has a repeated stripe pattern at regular spacing.
Repeating Patterns as Spatial Frequencies explained visually
The scan has a repeated stripe pattern at regular spacing. English captions are optional and start off. The complete transcript follows the written lesson.
What you will learn
The scan has a repeated stripe pattern at regular spacing.
A practical question
A laboratory scan of a metal surface shows pale stripes that repeat across the picture. There are too many stripes to describe one by one, so let us ask a simpler question: how often does the pattern repeat as we move across the image? A broad stripe pattern changes slowly. A tightly packed one changes quickly. That rate of change across image position is spatial frequency. It describes the image, not the physical source of the stripes.
The useful idea
Direction matters just as much as spacing. Vertical stripes change when we travel from left to right, but their brightness can stay nearly constant when we travel along them. Horizontal stripes do the opposite. If someone says only that the scan has a high-frequency pattern, we still do not know which way it runs. That detail matters later, because a filter aimed at one direction can affect different structures from a filter aimed at another.
Work through the example
Here is a small comparison you can make on paper. Draw three broad bands across a fixed-width rectangle. Beside it, draw twelve narrower bands across a rectangle of the same width. The second drawing has four times as many cycles across that width. Now rotate both drawings and ask which direction you would count in. In a real scan, repeat the comparison on a known region and keep track of its pixel dimensions. Counting in a cropped or resized copy without recording the scale can change the number you report.
Limits and next step
There is one important trap. A specimen might genuinely contain a repeated fine texture, while a scanner adds regular interference over it. Both appear as repeating patterns. A frequency measurement cannot, by itself, tell you which one belongs to the object. Compare the untouched scan with another capture or a trusted reference before deciding to remove anything. In the next lesson, we will look at a Fourier display and see how the spacing and direction of these stripes appear there.
Worked example
Draw three broad bands and twelve fine bands across equal-width grids. Count cycles in the direction where brightness changes, then rotate the grids and compare the direction of change.
Quick check: test your understanding
Try it yourself: On two equal-width scans, one pattern repeats four times and the other repeats sixteen times. Which changes faster across the width, and what extra evidence would you need before calling either pattern scanner interference?
Check the reasoning
The sixteen-cycle pattern has four times the spatial frequency across that width. Frequency alone cannot identify the cause. Compare raw captures, specimen detail, and a reference before treating the pattern as interference.
What is the main idea behind repeating patterns as spatial frequencies?
Higher spatial frequency means faster change over image position; orientation matters in two dimensions.
How is it used in the example?
Count stripe cycles across the image and compare a coarse pattern with a fine one.
What limitation should be checked?
A frequency describes repetition, not necessarily the physical cause of a stripe.
Further reading
Digital Image Processing, 4th edition — Rafael C. Gonzalez and Richard E. Woods.
Original ToolsInVision teaching examples. Editorial approach · Suggest a correction
Accessible referenceLesson 37 transcript322 words · Video transcript · Matches the published narration and captionsOpen transcript
A laboratory scan of a metal surface shows pale stripes that repeat across the picture. There are too many stripes to describe one by one, so let us ask a simpler question: how often does the pattern repeat as we move across the image? A broad stripe pattern changes slowly. A tightly packed one changes quickly. That rate of change across image position is spatial frequency. It describes the image, not the physical source of the stripes.
Direction matters just as much as spacing. Vertical stripes change when we travel from left to right, but their brightness can stay nearly constant when we travel along them. Horizontal stripes do the opposite. If someone says only that the scan has a high-frequency pattern, we still do not know which way it runs. That detail matters later, because a filter aimed at one direction can affect different structures from a filter aimed at another.
Here is a small comparison you can make on paper. Draw three broad bands across a fixed-width rectangle. Beside it, draw twelve narrower bands across a rectangle of the same width. The second drawing has four times as many cycles across that width. Now rotate both drawings and ask which direction you would count in. In a real scan, repeat the comparison on a known region and keep track of its pixel dimensions. Counting in a cropped or resized copy without recording the scale can change the number you report.
There is one important trap. A specimen might genuinely contain a repeated fine texture, while a scanner adds regular interference over it. Both appear as repeating patterns. A frequency measurement cannot, by itself, tell you which one belongs to the object. Compare the untouched scan with another capture or a trusted reference before deciding to remove anything. In the next lesson, we will look at a Fourier display and see how the spacing and direction of these stripes appear there.