Does the pho­to look dirty and fuzzy? Feel­ing like the image is filled with noise, but you can’t see the details? Does an old film effect appear that you did not plan? Most like­ly, we are talk­ing about a defect called noise.

We fig­ure out what it is, why there is noise in the pho­to and — most impor­tant­ly — how to remove it.

Noise visu­al­ly makes the pho­to dirt­i­er / Pho­to: Eliza­ve­ta Chechevit­sa / vk.com/lizma

What is noise and why does it occur

Noise is a defect in pho­tog­ra­phy. It man­i­fests itself in the fact that small ran­dom­ly scat­tered pix­els appear in the pic­ture. Because of this, where the sur­face should be smooth, mono­phon­ic, or there should be empti­ness at all (for exam­ple, the night sky), there is a feel­ing of “rough­ness”, fuzzi­ness. It shows up espe­cial­ly well when an ini­tial­ly dark frame is heav­i­ly light­ened in post-pro­cess­ing.

Noise is of two types:

1. Col­ored. In this case, the pix­els scat­tered over the image are col­ored in dif­fer­ent col­ors.

Green and pur­ple spots in the image are an exam­ple of col­or noise / Illus­tra­tion: Eliza­ve­ta Chechevit­sa, Photosklad.Expert

The eas­i­est way to remove such noise is to dis­col­or it in a spe­cial way in a graph­ic edi­tor in a cou­ple of sec­onds.

2. Bright­ness.

Lumi­nance noise appears as scat­tered pix­els of dif­fer­ent bright­ness — they can be light, almost white, gray, dark / Illus­tra­tion: Eliza­ve­ta Chechevit­sa, Photosklad.Expert

If you reduce this noise in the pho­to, it will inevitably lose sharp­ness. This is due to the fact that lumi­nance noise is removed by blur­ring the image. There­fore, it is believed that such noise is more dif­fi­cult to remove, and some­times even impos­si­ble. Nat­u­ral­ly, if you do not want to lose qual­i­ty.

Some­times it’s eas­i­er to pre­vent a prob­lem than to fix it lat­er. If you know why nois­es appear, you can take action in advance and reduce the chance of their occur­rence. Let’s take a look at the most com­mon rea­sons.

Why there is noise in the photo

  • Too high ISO.

Noise occurs where there is not enough light. In this case, the pho­tog­ra­ph­er falls into the trap — he needs a lighter frame, but if he rais­es the ISO, the qual­i­ty will drop. You should either put up with this, or look after your­self with a fast lens.

  • The state of the cam­era matrix.

The matrix may be phys­i­cal­ly dam­aged. For exam­ple, it may have bro­ken pix­els. This is the name of “bro­ken” pix­els that do not trans­mit col­or or always shine with some one col­or — red, green, blue.

  • Matrix pix­el size.

The big­ger it is, the bet­ter. Often this val­ue is indi­cat­ed in the char­ac­ter­is­tics of the cam­era, if not, in this text we tell in detail how to cal­cu­late the phys­i­cal size of a pix­el our­selves.

  • Cam­era and lens­es designed for ama­teurs.

As a rule, the more expen­sive the cam­era, the larg­er the pix­el size, as well as the high­er the ISO set­ting that the cam­era allows you to set. Where an ama­teur cam­era feels inse­cure at ISO 800, anoth­er calm­ly shoots at ISO 1200–1600 with­out noise and loss of qual­i­ty.

Pro­fes­sion­al lens­es also pro­duce bet­ter image qual­i­ty than their bud­get coun­ter­parts. This is the case when you over­pay not for mar­ket­ing, but for bet­ter tech­nol­o­gy. In addi­tion, only in more expen­sive mod­els you can find tru­ly fast cf / 2.8 glass­es and below. The low­er this num­ber, the dark­er con­di­tions you can shoot with­out bump­ing up the ISO.

  • Wrong post-pro­cess­ing.
Pho­to after removal of col­or noise / Illus­tra­tion: Eliza­ve­ta Chechevit­sa, Photosklad.Expert

Noise may appear if in the graph­ics edi­tor:

  • raise the expo­sure too much;
  • rad­i­cal­ly bright­en dark areas;
  • increase sharp­ness;
  • increase the sat­u­ra­tion — this high­lights the col­or noise;
  • when edit­ing white bal­ance. For exam­ple, you want to remove cold blue from a pho­to and nat­u­ral­ly add yel­low for this. In this case, the yel­low noise that was in the pic­ture will inten­si­fy.

How to remove noise from a photo

We will look at the eas­i­est and fastest ways to remove noise from a pho­to using image edi­tors. But remem­ber two things:

  • it is not always pos­si­ble to remove noise com­plete­ly;
  • get ready that the sup­pres­sion of lumi­nance noise will “kill” the sharp­ness of the image.

How to Remove Noise from a Photo in Photoshop

1. Open a pho­to and make a dupli­cate of the orig­i­nal lay­er. To do this, right-click and search for the com­mand Dupli­cate lay­er / Dupli­cate lay­er (Hotkeys Ctrl+J).

2. Stand on the dupli­cate lay­er by click­ing on it. Go to menu Fil­ter / Fil­ter — Cam­era RAW fil­ter / Cam­era Raw Fil­ter (Hotkeys Shift+Ctrl+A).

You will be in the Adobe Cam­era Raw mod­ule. It is need­ed for devel­op­ing pic­tures in RAW for­mat, but it also allows you to work with files in JPEG for­mat. Cam­era Raw has an excel­lent built-in noise reduc­tion, so there is no need to down­load or buy addi­tion­al plug-ins / Illus­tra­tion: Eliza­ve­ta Lentchevicha, Photosklad.Expert

3. Find a tab Detail and two groups of slid­ers in it: Noise Removal / ND and Col­or Noise Removal / Col­or ND. The first group is respon­si­ble for the lumi­nance noise that cre­ates grain­i­ness, and the sec­ond group is respon­si­ble for the col­or noise.

Menu for remov­ing col­or and lumi­nance noise / Illus­tra­tion: Eliza­ve­ta Lentchevicha, Fotosklad.Expert

Both groups have three slid­ers:

one. Noise Removal / ND and Col­or Noise Removal / Col­or ND.

Lumi­nance noise is removed by blur­ring, and col­or noise is removed by desat­u­ra­tion and col­or blur. This is the main noise removal slid­er.

2. Detail / Detail.

Attempts to main­tain the sharp­ness that the pre­vi­ous slid­er removed. The larg­er the val­ue of the slid­er, the more noise will be returned to the image.

In a group Col­or Noise Removal / Col­or ND this slid­er pro­tects col­or tran­si­tions. This is nec­es­sary so that the dis­col­oration con­cerns only noise and does not creep onto the mod­el, clothes, inte­ri­or.

3. Con­trast / Con­trast.

The high­er the val­ue, the high­er the image con­trast that the first slid­er could remove. High val­ues ​​can show dirt in the pic­ture — spots, grain­i­ness.

How to Remove Noise in Lightroom

Noise removal in Light­room is very sim­i­lar to the same algo­rithm in Pho­to­shop. The fact is that Light­room is a com­plete ana­log of Adobe Cam­era Raw, but only with a slight­ly mod­i­fied inter­face.

Noise reduc­tion pan­el in Light­room, as well as the result before-after

1. Open a snap­shot and go to the tab Devel­op­ment / Devel­op.

2. In the set­tings pan­el on the right, find the tab Detail / Detail.

3. In the Detail tab, find the col­umn Noise Reduc­tion / Noise Reduc­tion.

Slid­er Bright­ness / Lumi­nance respon­si­ble for lumi­nance noise, and Col­or — for col­ored ones. These are the basic tools for noise reduc­tion.

slid­ers Detailas in Pho­to­shop, set the thresh­old for the effect, bring­ing back detail and—often—noise too.

Slid­er Con­trast / Con­trast returns the con­trast of the image, and Smooth­ness / Smooth­ness for col­ored noise sets the smooth­ness of the tran­si­tion between col­ors, does not allow the tool to dis­col­or third-par­ty objects, except for noise.

How to take photos without noise — 5 tips

Pho­to before (left) and after (right) noise reduc­tion / Illus­tra­tion: Eliza­ve­ta Chechevit­sa, Fotosklad.Expert

1. Increase your shut­ter speed instead of ISO and shoot from a tri­pod.

2. Use addi­tion­al light if pos­si­ble. Even the cheap­est exter­nal flash on the report will raise the qual­i­ty of the pic­tures and reduce the noise.

3. Keep a fast lens (f / 2.8 or less) in your arse­nal in case of shoot­ing in poor light. For exam­ple, in con­cert halls, the­aters, it is for­bid­den to use a flash so as not to dis­tract the artists.

4. When choos­ing a cam­era, give pref­er­ence to the one with a larg­er pix­el matrix. For exam­ple, a Nikon D610 cam­era with a 35.9 x 24 mm sen­sor or a Canon EOS R6 mir­ror­less cam­era with a 35.9 x 23.9 mm sen­sor.

5. Shoot in RAW, not JPEG. In post-pro­cess­ing, it will be eas­i­er to “stretch” the image with­out los­ing qual­i­ty.