Outcome 6.2 Scanning: Resolution, Halftones, and File Preparation for Print

Understanding the relationship between LPI and DPI (and where scanning fits)

When you scan an image for print, you’re converting a physical original (a photo, drawing, or piece of text) into a grid of pixels. But printing doesn’t place “pixels” on paper—most printing processes simulate tones using patterns of tiny dots. That’s why lines per inch (LPI) and dots per inch (DPI) are linked: they describe two different layers of how printed tone is produced.

What LPI means (halftone screen frequency)

Lines per inch (LPI) describes the frequency of a halftone screen in traditional amplitude-modulated (AM) halftone printing (common in offset lithography and many print workflows). A halftone screen breaks an image into a grid of halftone cells. Each cell contains dots whose size changes to simulate light and dark areas.

• Higher LPI means smaller halftone cells and typically finer detail (up to the limits of paper, press, and ink spread).
• Lower LPI means larger cells, which can look grainier but can hold up better on rough paper where fine dots would fill in.

Why it matters: LPI is one of the main print-quality constraints. Even if your scan is extremely high resolution, the screen frequency and printing conditions may limit how much detail can actually appear on paper.

What DPI means (device dot placement)

Dots per inch (DPI) describes how many individual dots a device can place in one inch. In printing, it refers to the output device’s addressability (how finely it can position ink/toner dots). In scanning, people sometimes say DPI when they really mean scanning resolution, but for scanners the more precise term is often samples per inch (SPI) or simply scanning resolution in pixels per inch (PPI).

Why it matters: In halftone printing, the printer needs enough DPI to build each halftone cell out of multiple smaller dots. If the printer’s DPI is too low relative to LPI, you get poor tonal smoothness, banding, or screen artifacts.

The key relationship: halftone cells are built from printer dots

Think of it like mosaic art:

• LPI is like the size of each mosaic “tile” (halftone cell).
• DPI is like how many tiny pieces you can fit inside each tile.

A simplified way to connect them (for AM halftone screening) is to look at how many printer dots fit across a halftone cell:

$\text{dots per cell side} = \frac{\text{DPI}}{\text{LPI}}$

If a cell is (roughly) a square, the total dots in the cell is approximately:

$\text{dots per cell} = \left(\frac{\text{DPI}}{\text{LPI}}\right)^2$

More dots per cell generally allows smoother tonal transitions (more possible dot sizes). In a basic model of AM halftoning, the number of distinct tonal steps a single cell can represent is approximately the number of dot-on counts plus one:

$\text{tone steps} \approx \left(\frac{\text{DPI}}{\text{LPI}}\right)^2 + 1$

This model is a teaching approximation—real screening algorithms and press behavior complicate things—but it captures the essential idea: higher DPI relative to LPI supports smoother tone.

Where image resolution (PPI) connects to LPI

For preparing images for halftone printing, a widely used guideline is:

• Image resolution (PPI) should be about 2× the LPI at final print size.

So if a job prints at $150\,\text{LPI}$, a common target is about $300\,\text{PPI}$ for the image at the final reproduction size.

This is about sampling theory and practical tradeoffs: too low PPI risks visible pixelation or loss of detail; far beyond what the halftone screen can use tends to bloat files without visible improvement.

Putting the terms in the right boxes (common confusion)

A big source of mistakes is mixing up what each unit describes.

A helpful memory aid: Pixels (PPI) are in your file, Dots (DPI) are in your device, Lines (LPI) are in your halftone screen.

Example: relating LPI to image PPI

If a print shop specifies $150\,\text{LPI}$, you typically aim for:

$\text{image PPI} \approx 2 \times 150 = 300\,\text{PPI}$

That doesn’t mean your printer is $300\,\text{DPI}$. Many workflows image plates or print at much higher device resolution (often in the thousands of DPI) to construct clean halftone dots.

What goes wrong in practice

• If you scan extremely high “DPI” but then place the image very small, you might be carrying useless data.
• If you scan too low, then enlarge the image in layout, the effective PPI drops and the print can look soft or jagged.
• If you confuse DPI (printer) with PPI (image), you might set the wrong targets entirely.

Exam Focus

• Typical question patterns:
  • Explain the difference between LPI, DPI, and PPI, and describe how they interact in printing.
  • Given an LPI, choose an appropriate image PPI for print reproduction.
  • Conceptually describe why a printer needs higher DPI than the halftone LPI.
• Common mistakes:
  • Treating DPI, PPI, and LPI as interchangeable terms.
  • Saying “scan at the printer DPI” instead of basing scan resolution on final image size and required PPI.
  • Ignoring that LPI is tied to printing conditions (paper, press), not just “higher is always better.”

How resolution affects file size (and why “higher” isn’t always better)

Resolution in scanning and imaging usually means how many pixels you capture—either stated directly as pixel dimensions (e.g., $4000 \times 6000$ pixels) or as PPI relative to a physical size. File size is the amount of storage the image takes, which depends on how much data is stored per pixel and whether compression is used.

What “resolution” really changes: number of pixels

If you scan a physical original with width $W$ inches and height $H$ inches at $R$ pixels per inch, the pixel dimensions are:

$\text{pixels}_x = W \times R$
$\text{pixels}_y = H \times R$

Total pixel count is:

$\text{total pixels} = (W \times R)(H \times R) = WHR^2$

That squared term matters: doubling resolution in PPI quadruples the number of pixels (if the physical scan area stays the same).

Bit depth and color channels: how much data each pixel carries

Each pixel stores color information. How much depends on:

• Color mode / channels (e.g., grayscale has 1 channel; RGB has 3; CMYK has 4)
• Bit depth per channel (commonly 8 bits/channel or 16 bits/channel)

For an uncompressed raster image, a practical estimate of file size is:

$\text{file size (bytes)} \approx (\text{pixels}_x \times \text{pixels}_y) \times \frac{\text{bits per pixel}}{8}$

Where:

$\text{bits per pixel} = (\text{channels}) \times (\text{bits per channel})$

This gives a baseline before compression and before extra features like layers, masks, alpha channels, or embedded previews.

Compression: why two files with the same resolution may have different sizes

Many formats compress data:

• Lossless compression (e.g., TIFF with LZW/ZIP, PNG) reduces size without changing pixel values.
• Lossy compression (e.g., JPEG) reduces size by discarding some information, which can introduce artifacts—especially around sharp edges and text.

So resolution strongly influences file size, but it’s not the only factor. Two images with identical pixel dimensions can vary greatly in file size due to compression level and image content (busy textures compress differently than flat areas).

Example 1: doubling PPI increases file size about 4× (same scan area, same bit depth)

Suppose you scan a $5\,\text{in} \times 7\,\text{in}$ photo.

• At $300\,\text{PPI}$:
  • $\text{pixels}_x = 5 \times 300 = 1500$
  • $\text{pixels}_y = 7 \times 300 = 2100$
  • Total pixels $= 1500 \times 2100 = 3{,}150{,}000$

If it’s RGB at 8 bits/channel, then bits per pixel is:

$\text{bits per pixel} = 3 \times 8 = 24$

Uncompressed size is:

$3{,}150{,}000 \times \frac{24}{8} = 3{,}150{,}000 \times 3 = 9{,}450{,}000\,\text{bytes}$

That’s about $9.45\,\text{MB}$ (using $1\,\text{MB} = 1{,}000{,}000\,\text{bytes}$ as a simple estimate).

• At $600\,\text{PPI}$ (double the PPI): pixel dimensions double in each direction, so total pixels quadruple:
  • Uncompressed size becomes about $4 \times 9.45\,\text{MB} \approx 37.8\,\text{MB}$

The key learning: increasing PPI increases file size very quickly.

Example 2: changing “PPI” without resampling does not change file size

If you take an image that is $3000 \times 2400$ pixels and simply change its PPI metadata from $300\,\text{PPI}$ to $150\,\text{PPI}$ without resampling, the pixel dimensions stay the same—so the file size stays essentially the same. Only the implied print size changes.

This is a frequent trap: students think “lowering resolution” always reduces file size. It only reduces file size if you discard pixels (resample down) or apply stronger compression.

What goes wrong in practice

• Scanning at extremely high resolution “just in case” can create huge files that slow down editing and layout—and may not improve print quality.
• Saving a scan as JPEG too early can bake in artifacts that become obvious after retouching or text overlay.
• Confusing effective resolution (PPI at placed size in layout) with file resolution metadata can lead to unexpected soft prints.

Exam Focus

• Typical question patterns:
  • Predict how file size changes if you double the scanning resolution (same physical dimensions, same mode/bit depth).
  • Calculate approximate uncompressed file size given pixel dimensions and bit depth.
  • Explain why two images with the same resolution can have different file sizes (compression, bit depth, layers).
• Common mistakes:
  • Assuming file size scales linearly with PPI (it scales roughly with $R^2$ for the same physical scan area).
  • Thinking changing PPI metadata alone changes file size (it doesn’t unless you resample).
  • Ignoring bit depth and channels (grayscale 8-bit is much smaller than RGB 16-bit at the same pixel dimensions).

Scanning workflow skills: input controls, calibration, formats, and scaling to the right resolution

Scanning for digital print design is less about “pressing scan” and more about controlling the chain from physical original to printable file. You want predictable color, correct tone, minimal artifacts, and the right resolution at the final reproduction size.

Manipulating input functions (scanner settings you actually control)

A scanner’s software typically offers many controls. The important idea is that each control trades off quality, flexibility, and file size.

1) Optical resolution vs interpolated resolution

On scanner spec sheets you may see very high numbers, but only optical resolution reflects the true detail the hardware can capture. Interpolated resolution is software-generated pixels guessed from neighbors.

Why it matters: Interpolation increases file size but rarely adds real detail. For print-quality work, prioritize optical resolution and only upsample later if you have a specific need.

2) Color mode and bit depth

You usually choose:

• Grayscale for black-and-white photos or pencil drawings where color isn’t needed.
• RGB color for color originals (most scanning workflows capture in RGB first, even if the final output is CMYK).
• Black-and-white / line art (1-bit) for pure text or inked line drawings.

Bit depth choices commonly include:

• 8 bits/channel (good general-purpose, smaller files)
• 16 bits/channel (more editing headroom for heavy corrections, larger files)

Why it matters: Higher bit depth helps avoid banding and preserves subtle tonal transitions during editing—but it increases file size and may be unnecessary for simple jobs.

3) Exposure/levels controls: black point, white point, and midtones

Scanning software may offer histogram/levels tools. Setting these correctly:

• prevents “flat” scans (no true blacks/whites)
• avoids clipping detail in shadows/highlights

Common mistake: Setting black/white points too aggressively during scanning. That can permanently clip detail you could have recovered later. For important originals, it’s often safer to capture a slightly “flat but complete” scan and refine in an editor.

4) Descreening and moiré control

If you scan something that was already printed (a magazine photo, brochure, newspaper), you may capture the halftone dot pattern. When that pattern interacts with the scanner sampling grid, you can get moiré—wavy interference patterns.

To manage this, many scanning tools offer descreen options. You can also reduce moiré by scanning at a higher resolution and then downsampling carefully, but descreening is often the first tool to try.

5) Sharpening and “auto enhance” features

Many scanner drivers apply automatic sharpening, contrast boosts, or color enhancements.

Why it matters: These can look pleasing immediately, but they reduce your control and can create halos or exaggerated noise. In professional print workflows, you often disable heavy “auto” effects, capture a clean scan, and apply controlled sharpening later.

Calibrating scanning equipment (getting predictable color and tone)

Calibration means bringing a device into a known, repeatable state. For scanning, the goal is that the same original scanned today and next week looks consistent—and that what you see on screen is a reliable guide for editing.

Scanner calibration and ICC profiles (conceptual overview)

Color-managed workflows rely on ICC profiles to describe how a device reproduces color. A scanner profile characterizes how your scanner captures colors so software can correct them toward a standard color space.

A common professional approach uses a known reference target (often called an IT8 target in many workflows) and profiling software to build an ICC profile. Not every classroom or lab will do full profiling, but you should understand the purpose:

• reduce color casts
• improve consistency across devices
• make edits more predictable

Practical calibration habits you can control

Even without full profiling software, you can significantly improve scanning results by controlling fundamentals:

• Warm up the scanner if recommended by the manufacturer (some lighting systems stabilize after a short time).
• Clean the glass (dust creates repeated spots you’ll waste time retouching).
• Cover the scan bed to prevent stray light for reflective scans.
• For film/transparent scanning, keep holders and film clean to reduce dust and Newton rings.

Also, remember that scanning is only half the story: if your monitor is too bright or inaccurate, you’ll “correct” scans incorrectly. A well-calibrated display supports better scanning decisions.

Saving scans: choosing formats for print design

File format choice affects quality, editability, compatibility, and file size.

TIFF (common for print workflows)

TIFF is widely used for print because it can store high-quality raster data, often with optional lossless compression, and it’s broadly supported.

• Best when you want to preserve quality and avoid generational loss.
• Good for archiving a master scan.

JPEG (small, but lossy)

JPEG uses lossy compression.

• Best for final delivery when small size matters and slight loss is acceptable.
• Risky as a master format—each resave can degrade quality, and artifacts can become visible in print (especially around text, logos, and sharp edges).

PNG (lossless, common for screen; sometimes useful for line art)

PNG is lossless and handles flat-color graphics well. It’s more common in screen workflows but can be useful for certain scan types (like clean line art) when TIFF isn’t required.

PDF (container format)

PDF can contain images (and vectors/text). Some scanners output directly to PDF for document workflows, but for design work you typically want control over image editing before embedding in a PDF layout.

Native editor formats (PSD and similar)

If you edit heavily, a format like PSD can preserve layers and masks. You might keep:

• a master editable file (layered)
• a flattened delivery file (often TIFF or high-quality PDF, depending on the job)

Common mistake: Scanning straight to JPEG, then doing heavy retouching and resizing, then saving again as JPEG. That stacks compression artifacts.

Scaling, sizing, and adjusting resolution correctly

This is one of the most testable skills: you must control the relationship between pixels, physical size, and PPI.

The core relationship

For a given print size:

$\text{pixels} = \text{inches} \times \text{PPI}$

If you know the final reproduction size and the target print PPI, you can compute how many pixels you need.

Effective PPI in layout (why your scan can “lose resolution”)

If you place an image into a layout program and scale it up, the effective PPI decreases because the same pixels are spread over more inches. If you scale it down, effective PPI increases.

A practical relationship (without changing pixel count) is:

$\text{effective PPI} = \frac{\text{original PPI}}{\text{scale factor}}$

Where scale factor is expressed as a multiplier (e.g., 200% enlargement is factor $2$).

Resampling vs changing PPI metadata

• Changing PPI without resampling changes only the intended print size.
• Resampling changes the actual pixel dimensions (adding or removing pixels), which changes file size and can affect quality.

Common mistake: Students “set the resolution to 300 PPI” by typing 300 into a box while resampling is off, thinking they improved quality. If you didn’t add pixels, you didn’t add detail—you just told the software to print it smaller.

Example 1: choosing scan resolution based on enlargement

You have an original photo that is $4\,\text{in} \times 6\,\text{in}$. You want to print it at $8\,\text{in} \times 12\,\text{in}$ (a 2× enlargement in each dimension) at $300\,\text{PPI}$ effective.

Because you’re enlarging by a factor of $2$, you need to scan at:

$\text{scan PPI} = 300 \times 2 = 600\,\text{PPI}$

That way, when you enlarge the placed image to 200%, its effective PPI becomes approximately $300\,\text{PPI}$.

Example 2: scanning line art vs photos

If you scan:

• A continuous-tone photo: scanning around print targets (often ~300 PPI at final size) is common.
• Line art (pure black ink): you often need a much higher scan resolution to keep edges crisp because there are no “soft gradients” to hide jaggedness. This is why line art is frequently captured at higher PPI than photos in print workflows.

Instead of memorizing one magic number, think in terms of edge fidelity: sharp edges need more pixels per inch than smooth tones to look clean.

A step-by-step scanning workflow you can practice

1. Identify the source: photo, printed halftone, text document, line drawing.
2. Decide the final print size in the layout.
3. Choose a target effective PPI appropriate to the job (often guided by print specs).
4. Calculate scan PPI based on any enlargement/reduction.
5. Set scanner options:
   • color mode (RGB/grayscale/line art)
   • bit depth (8-bit vs 16-bit)
   • optical resolution (avoid “interpolated” settings)
   • descreen if scanning printed material
   • disable aggressive auto enhancements if you need control
6. Preview scan and adjust cropping, orientation, and tonal endpoints conservatively.
7. Scan and save a master (commonly TIFF or a layered format if editing is expected).
8. Edit and retouch (dust removal, color correction, controlled sharpening).
9. Export delivery versions appropriate to the workflow (TIFF/JPEG/PDF as required), embedding or preserving profiles as your workflow expects.

What goes wrong in practice

• Scanning a printed magazine photo without descreening often produces moiré—then students try to “fix” it with blur, losing detail.
• Cropping too loosely and planning to crop later wastes pixels and file size.
• Scanning in CMYK directly (when the scanner/software even allows it) can reduce flexibility; many workflows scan in RGB first and convert later under controlled settings.
• Using auto-sharpen can create halos that become obvious after resizing and printing.

Exam Focus

• Typical question patterns:
  • Given an original size and final size, calculate the scan PPI needed to achieve a required effective print PPI.
  • Identify which scanner settings help when scanning a previously printed image (moiré/descreen).
  • Choose an appropriate file format for a master scan vs a delivery file, and justify why.
• Common mistakes:
  • Confusing optical resolution with interpolated resolution and selecting inflated settings.
  • Forgetting that enlarging in layout reduces effective PPI (leading to soft prints).
  • Saving masters as JPEG and repeatedly resaving, causing visible compression artifacts in print.