How to increase iq as fast as possible every day

Help: Improve the quality of your pictures

Under work, help is always welcome

This page has as purpose to gather know-how and tricks in order to improve the quality of your photographs, both when taking them and when processing them.

Basics of photography - how you learn to shoot better photos

Before starting to shoot [edit]

Maintain your equipment clean [edit]

Keeping your equipment, especially the lenses clean, will save you a lot of time later on. Don't forget to clean your lenses using an air blower or cloth.

If you don't clean the lenses in a regular way you will spend much more time later one finding and cloning out the dust spots in your pictures. To avoid permanent damage it is recommendable to use an UV filter.

Camera configuration [edit]

Manual mode, 5 f-numbers lower than automatic.

Using the camera in automatic mode is comfortable and quick, but the results will often differ from the intention. As you can see in the 2 shown pictures in automatic mode the camera tries to balance light and shadows, which will often result in over or underexposed images when the light conditions are not standard. With a manual f / 8 and 1/500 s configuration the result is more impressive than in automatic (f / 2.8 1/320 s).

Changing the usage of the camera from automatic to manual will require additional effort and will initiate a new "learning curve" but after you have understood and dominated it, you will never want to go back to automatic. You will only have total control of your camera and so express what you want to in manual mode.

Another crucial setting is the file format. There are mostly 2 options: RAW and JPEG formats (a few cameras also support DNG, a kind of standard raw format among manufacturers). RAW gives you much more flexibility when you process your pictures and it is a kind of "second chance" to adjust the settings. You will be able to recover overexposed areas, lighten up darker areas, or even darken or brighten your pictures up to 2 f-numbers without sustantial loss of quality. Shooting in RAW has also some disadvantages: It requires more space on your memory card, you cannot use direct print functions of your camera, besides the DNG format the RAW formats of the different camera brands are incompatible with each other.

Making the picture [edit]

Choosing a subject [edit]

Before you get to a place you can check in related articles (e.g. in Wikipedia) whether these have pictures and whether all important subjects have been captured.

If you travel to a foreign country it is very important that you get to know first the local legislation regarding freedom of panorama. In some countries pictures of artworks like sculptures or pictures are not compatible with the Commons licenses, in others even making pictures of buildings is not aligned with Commons's licenses. For more information, see Commons: Freedom of panorama.

There are many questions about licenses, copyright and public domain. For additional information, see Commons: Licensing.

Choosing time and place [edit]

  • regard to avoid disturbing objects like cars in front of your main object
  • look out for optimal time shooting your object (position of the sun, avoid crowd, ...)
  • searching a good position

Framing [edit]

Bear in mind that all elements on the picture are relevant and play a role in the composition. The job is not done centering the main subject, but rather looking for a favorable angle and adding or removing additional elements to the composition.

A lot of pictures are not good due to a bad crop of the main subject or even background elements. If there is too much sky or water look for a different focus or additional elements to use as much as possible of the original shot.

Focusing [edit]

If not configured in a different way, the focus is usually in the middle of the frame, but it doesn't necessarily mean that the focused area has to be in the center. Therefore, the framing shall be adjusted after the focus has been performed keeping the release pressed halfway.

When you observe a subject the brain is capable of blending out the background, but later on it can turn out disturbing, i.e. Branches or antennes that come out of the head of the main subject or just a busy background that can hardly be made out of the main subject. This kind of problems can be fixed looking for a new camera position or a different focal length in order to make the background blurry.

Be careful with the sun or strong light sources that can cause flares. To avoid them look for a different angle or use an object (e.g.) or your hand to avoid direct exposure of the sun.

Further suggestions [edit]

  • Applying the rule of thirds to improve the composition of your images.
  • Look for symmetry in your pictures (in opposition to the rule of thirds) using water, glass or mirror reflexes or repetitive elements.
  • In case of unusual big or small subjects, it can be useful to provide a hint about its size (adding e.g. an element of well known size like a person).
  • Be careful with contre-jour scenes. They emphasize the outlines of the elements but loosing detail of them.

Light [edit]

Interior [edit]

Photographs without flash keep the original atmosphere.

When making pictures withough flash in interiors the existing light sources will influence the result. Many of these sources result in warm colors, which could transmit a nice atmosphere but at the same time an unreal one.

Sunlight [edit]

Photography under bad weather conditions shouldn't be a problem, but rather an opportunity to take great shots.

Under sunlight, or even with cloudy sky, it is usually easier to make high quality pictures. Under rainy conditions the colors are more saturated and note eroded by the sun.

On the other side, a strong sun can cause harsh shadows, an especially remarkable problem of portraits. The harsh shadows could also result in a loss of detail of the underexposed areas

  • Balanced lights and shadows

  • Disturbing hard shadows with sunlight

Night photography [edit]

Night photograph half an hour before sunrise.
Night shot with strong light-dark contrast.
Night picture with remaining light, midsommer in in Helsinki.

Pictures taken in the night or under poor lighting require a slower shutter speed to allow the reception of mor light. This could begin with 1/10 seconds up to several minutes. Due to these long exposure time a tripod or stable base is required.

Apart from the long exposure time needed a second problem of this type of photography is the extreme contrast between very bright light sources and the darkness of the night. This is the reason why the HDR (Hight Dynamic Range) technics is used often for this purpose.

In particular, digital cameras quickly reach their limits when it comes to high contrasts, and the exposure time and aperture have to be chosen very carefully. It is advisable to take several photos with different exposure so that you can later select the best one or even combine several to form one image.

The best night shots are not taken in complete darkness, but just before sunrise or just after sunset during the so-called blue hour. Due to the lighting that is already or still present, the scene looks like a night shot, but the remaining light of the twilight also makes the otherwise unlit parts of the subject visible.

The Schwarzschild effect must be taken into account when shooting at night. In simple terms, this means that the exposure metering by cameras or exposure meters no longer works properly from exposures of approx. One second, i.e. the exposure time must be extended manually. Since there is currently no reliable mathematical model for this, a series of exposures with different extension factors is recommended. As a very rough rule of thumb you can expose for 15 seconds with a measured time of 10 seconds and 120 seconds with a measured time of 60 seconds. The degree of elongation depends heavily on the film material used.

This effect does not occur in digital photography because the CCD chips or CMOS image sensors used do not lose their sensitivity. However, other effects are noticeable in the long-term recordings, e.g. B. overexposure ("eating away") or dark noise, which is hardly of any consequence in modern SLR cameras.

Night shots appear warmer in the final image on conventional film if manual white balance is not performed. This is due to the color temperature of artificial light sources, which is lower than that of the sun or flash lights. Within certain limits, this can be compensated for in image processing.

If the camera used allows long exposure times, night shots are relatively problem-free as long as a tripod is used or the camera can be placed on any object. It is recommended to use the self-timer, because you can then hold the camera with both hands.

The halo effect often occurs in night shots. This means that light sources "outshine", that is, appear larger than in reality. This is normal and can hardly be avoided, but it is not a shortcoming.

Long exposures [edit]

Night shot with car lights

Effects such as lines of light from moving vehicles only succeed with exposure times of several seconds, depending on the distance and speed. With such long exposure times, the Schwarzschild effect occurs when taking pictures on conventional film. This is not present in digital photography, but you get unwanted noise in the image, which is roughly comparable to increased film grain.

In order to consciously achieve long exposure times, it is necessary with most modern cameras to switch off automatic programs, flash or motif programs. Instead, you can choose an automatic shutter speed or automatic shutter function, or you can expose with a manual setting. By choosing the smallest possible aperture (in the example: 22) or using a neutral density filter, the camera is forced to expose for a long time. The other way around, you can set a long time and leave the aperture selection to the automatic camera. An overexposure of up to three f-stops is - in contrast to daylight photos - possible without any problems.

If, as in the example, car headlights, blue lights, etc. are to appear in the image as a series of lights, the exposure measurement must be carried out at a time when there are no vehicles in the image. For long exposures, we recommend using a tripod and a wired or remote release or the self-timer, in the case of SLR cameras with the mirror folded up beforehand, in order to prevent vibrations in the camera.

Lightning [edit]

The flash units built into many cameras are only suitable to a limited extent for taking good pictures. Even with expensive cameras, the range is usually limited to just three to six meters. These flashes also create hard shadows, and the close proximity to the lens increases the risk of the notorious red-eye effect when taking photos of people.

As a result, objects and people close by are often too bright, and the rear area of ​​the image too dark. Post processing does not help much here either. A more skilful selection of the cutout can help; in the present example the built-in flash was not sufficient because the threshing machine was too far away.

If you get stuck without the flash, we recommend a device with a swivel reflector for indirect flash. Integrated flash units, on the other hand, often provide shading when taking close-up photos.

Flash light is also prohibited

  • when people can be disturbed by lightning, for example in concerts. Oversized external flash units can even cause permanent eye damage by bystanders looking in the direction of the flash.
  • when light-sensitive materials can be damaged, for example in a museum. In museums or similar interiors, you should only use the flash after consultation.

Back light

View from the bridge to Seurasaari Island (Helsinki)

Automatic programs of cameras rarely work satisfactorily for backlit shots. The automatically exposed images usually become too dark if the light source (s) are in the measuring range. If the camera allows, multiple spot metering on areas that are important to the image, the calculation of an average value and manual adjustment of the camera is recommended. In the right square image, the right and left image area was measured with an EOS-5 with 8 - 1/350, the reflections in the water with 11 - 1/500 and the camera then set to 22 - 1/60 because water at one exposure time 1/60 looks the most realistic.

If the backlight situation is such that there is only one object in a short distance, it can also be useful to switch on the flash to brighten the foreground compared to the background and thereby reduce the range of contrast somewhat.

The film or sensor sensitivity [edit]

The film speed is printed legibly on the cartridges (red arrow) and attached as so-called 'DX coding' (blue arrow) so that it is legible for the camera.

The ISO value (also ASA value) indicates the sensitivity of the film material or the sensor of the digital camera. With analog cameras, this can be selected by selecting the appropriate film. With digital cameras this can be set - depending on the model in the range between ISO 50 and ISO 6400 - or it can be selected fully automatically. Many simple viewfinder cameras and cheap SLR cameras can only recognize films with 100 to 400 ASA and switch to 100 ASA for unknown values, which causes incorrect exposure. If in doubt, you should consult the operating instructions or choose films with 100, 200 or 400 ASA.

The higher the ISO value, the greater the graininess or the image noise, but the shorter the exposure time can be with the same aperture. Conversely, with the same exposure time, a smaller aperture can be selected and thus a greater depth of field can be generated. Values ​​greater than ISO 200 to 400 should only be used if the situation makes it absolutely necessary. Many compact digital cameras have very small sensors that already have a disturbing noise at ISO 200 and deliver almost unusable results at ISO 400 and more. But even with higher quality digital cameras you should prefer low ISO settings if possible.

Sharpness [edit]

Images should be sharp because there is more detail in a sharp image than in an out of focus image. A distinction is made between motion blur and camera shake. With the former, the object moves during the exposure, with the latter, the photographer moves. Shake can be prevented by a short exposure time. As a rule of thumb, this time should not be longer than the reciprocal of the focal length for handheld shots. A picture with a focal length of 100 mm should therefore not be exposed freehand for longer than 1/100 of a second. This refers to the 35mm format. With digital cameras, the exposure time must be reduced by the format factor. An alternative to the short exposure times is to use a tripod.

Depth of field [edit]

Shallow depth of field and therefore a blurred background

The depth of field (colloquially often called depth of field) indicates the area that is still in focus in front of and behind the focused point. A high depth of field is often useful in documenting photography because so many parts of the object can be easily recognized.

The depth of field depends on the image scale and the set aperture. In practice, a large depth of field can be achieved with a small aperture (= large aperture number, for example f / 22) and with a large distance from an object compared to the depth.

A shallow depth of field is useful if you want the object to stand out clearly from the background and cannot achieve this in any other way. A shallow depth of field is therefore a popular stylistic device in portrait photography, for example, in which the outline of a head should not merge smoothly into the background, but rather stand out from it. A small depth of field can be achieved with a large aperture (= small aperture value, for example f / 1.4) and with a small distance from an object compared to the depth.

Aperture and exposure time [edit]

Closed aperture (22 left), aperture 11 (center) and open aperture (3.5 right) in comparison

The diaphragm is a construction located in the lens that forms a hole that can be changed in size. Together with the exposure time, it determines the exposure of the film or sensor. While the shutter speed influences camera shake and motion blur, the aperture determines the depth of field.

Relationships under given lighting conditions:

Almost all lenses on the market, regardless of whether they are for the extremely expensive SLR or for the simple digital camera, decrease in imaging performance when fully opened. On the other hand, very small aperture openings ensure increasing blurring due to diffraction effects. If the subject, light conditions and setting options of the camera allow it, one should always try to set medium aperture values ​​for optimal sharpness in documentary photography. In poor lighting conditions, a tripod is often the better alternative to using the flash.

Post-processing [edit]

Each processing step can destroy image information, so that post-processing - from a physical point of view - usually means “deterioration”.Therefore, you should never edit the original file of the photo, always only a copy. Image editing software is often supplied with the digital camera, and there is also free image editing software such as Gimp. Finally, Wikipedia also helps: picture workshop.

Typical beginners' mistakes have been eliminated in the example photos on the right: tonal value correction, dirt on the windows removed, slight sharpening and detail enlargement - an average picture has already become something worth seeing. When taking pictures from the car, the camera should always be held very close to the window, this prevents the dirt on the window from being depicted.

Sort out images [edit]

After taking the photos, you sort out the good pictures from the bad ones. Many pictures fail (motifs double or similar, person looks "stupid", hand in the picture, out of focus, poor section, people cropped, too light / dark etc.). The "reject rate" can be very different, but is often 50% to 75%, and for professional photographers even higher. Remember: Every "bad" picture damages the overall impression of all pictures. Ultimately, further processing is only worthwhile for recordings that have at least a certain minimum quality.

Select section [edit]

The example shows how important the image section is for the overall impression. Often disturbing elements push themselves into the picture, which can be removed with the detail processing and increase the value of the picture. Depending on the representation, such elements can also be desirable in order to create a three-dimensionality.

Optimize brightness distribution [edit]

In the example, the sky and the rest of the image were edited separately (with a 12px soft selection edge). If you were to brighten the above picture as a whole, the result would be a sky that is too bright. Almost all digital cameras produce overexposed images in slightly problematic lighting conditions; this is automatically corrected in laboratories during exposure. Only cameras far beyond 1000 euros do not have this unpleasant feature. If you have mastered these techniques, you should be in image processing Not work with the menu item Lighten, but instead carry out a channel-separated tonal value correction. With certain (mostly yellowish) motifs, it can make sense to convert the image to CMYK for this step. Almost all tone value curves have gaps in the highlights or shadows, sometimes only in individual channels. That's why you should post-process every digitally generated photo in this way. This problem almost never occurs with analog images that have been scanned (regardless of whether they are from negative or paper).

Optimize contrast [edit]

The selection of the appropriate contrast depends on the photographed object and the objective of the representation. Pictures taken in low light can add value if the contrast is increased and details can be seen more clearly. When taking portraits, some details such as unevenness and pores in the skin, which can be resolved by reducing the contrast, are annoying.

Image size [edit]

Using a camera with several million pixels does not automatically guarantee a sharp and good picture. However, this gives you "reserves" which can be useful for excerpts, for example.

Arguments for large images with little compression [edit]

Since lossy compression is carried out when creating JPEG images, i.e. information is reduced, which leads to a considerable loss of quality when used repeatedly (e.g. during image post-processing), it is advisable to keep the JPEG quality not below 0, when saving. 90 to put. If image post-processing by third parties is likely (or even explicitly desired; wiki principle!), Because e.g. B. excerpts, color values ​​and noise / sharpness still need reworking, then a JPEG quality of 0.95 is recommended. Otherwise, the compression artifacts show through to revisions except for the preview view ("thumb"). This effect occurs in particular when a rotation is necessary for architectural photographs ("Crooked horizon") or converging lines / perspective distortions ("Buildings too close") are to be corrected. In such cases, the editors from the photo workshop are happy if they don't have to bother with raw material in 800 pixels width, which suddenly looks completely blurred after turning and rectification and only looks unnatural when resharpened.

It is advisable to leave the image unchanged, particularly in the case of images of high encyclopedic value, which are, however, afflicted with photo-technical deficiencies as it comes from the camera (without any corrective action) and ask for help in the picture workshop. The longer times for uploading should be accepted.

Optimize colors [edit]

Post-processing is necessary with all digital cameras if the lighting conditions were not ideal when the picture was taken. This works well with the "tonal value correction" function in image processing. In any case, this should be carried out separately for each channel. If the result is unsatisfactory, which is often the case with a yellow cast, converting the image to CMYK can help.

Retouching [edit]

Retouched image and original

Smaller retouching can be used to improve an image, for example by making smaller, annoying objects, shadows, dirt, etc. disappear in order to better bring out the main subject. With an encyclopedia like Wikipedia, however, it is important that the actual image content is not falsified.

Disturbing details such as a power line in the background that appear to be growing out of a head should be tried to remove. Such retouching takes some practice. Sometimes you can only see retouching if you know the original (see example - bicycle).

Example of extensive retouching

It is best to pay attention to disturbing things while taking pictures so that you don't have to do a lot of post-processing in the first place. This is not always possible, as the second example shows. The photo served as the basis of a concept for redesigning the outdoor advertising of a shop as part of a master's thesis. Therefore, all disturbing things such as cars, bicycles, traffic signs, street lights with signs and shadows were subsequently removed. The work on the example took more than four hours.

Old photos are often damaged, wrinkled, dog-eared, or even missing parts. Here it is advisable to carefully touch up the corresponding areas. It is advisable to work at 200%, but to reduce the size every now and then in order to be able to assess the effect. Photos look more authentic when there are dog-ears, etc. The removal of image defects is not always the right way; a sure instinct is required here for the assessment.

Scaling [edit]

Scaling means changing the image resolution. It is possible to enlarge an image by calculating (interpolating) additional pixels, but the image does not gain in sharpness or detail, and no information is added. However, it often makes sense to first enlarge a template for the purpose of simpler or higher-quality post-processing, in order to then downgrade it to the original format. In the opposite case, by omitting pixels, the sharpness can be increased or it is no longer noticeable that the image is blurred.

Sharpening

Sharpness filters in image editing programs can seemingly eliminate slight blurring. However, these filters should be used with great caution and should always be the very last processing step on an image. Images for monitor use can be sharpened a little more than images that are exposed. When sharpening heavily compressed JPG files, compression artifacts appear ugly.

Image format and transparency [edit]

File format

  • At the recording:
    • Technically ideal is saving in RAW mode or in TIF format, if the camera allows this. Both methods save losslessly. While RAW always requires special post-processing on the computer, but offers extensive possibilities for influencing it, many parameters such as white balance are already specified by the camera in the TIF format. Disadvantages of these formats are the high memory consumption and, with some cameras, very slow storage.
    • Photos in JPG format should be taken at the highest quality setting. The lower data compression produces fewer artifacts, and the photos allow much better possibilities for post-processing. Once images are broken, they cannot be improved afterwards by any measure.
    • Images in jpg format can not only be broken if the compression is too high, but also simply by repeated overstoring. Every new saving of an image as a jpg file also triggers a new compression. During the image processing of a jpg image, care should therefore be taken to save intermediate steps only in a lossless format (tiff, eps, psd, png, bmp) and to only transfer the corrected file to a (new, not overstored) jpg file at the end. Format to convert.
  • Before uploading:
    • After completing all editing steps, photos should be scaled to a reasonable size and appropriately compressed. Many image edits allow a preview so that you can find a good compromise between file size and quality. It is advisable to save the version to be uploaded under a different name and to keep the original recording in order to be able to start over in the event of "processing accidents".
    • Drawings and screenshots should be saved in the lossless PNG format, since the abrupt color changes that are common here would result in a loss of quality even with high JPG quality settings. In particular, lettering in JPG format is often defaced by compression artifacts, and black and white images are also comparatively poorly compressed. The GIF format is considered obsolete because it can only display a maximum of 256 color gradations.
    • Animations can be saved in GIF format. The MNG format for animations is currently only supported by a few browsers and is therefore not recommended at the moment.

resolution

The resolution should be as high as possible, since Wikipedia articles also appear in printed form. However, the resolution should be appropriate for the object depicted and the quality of the recording. For example, it does not make sense to interpolate recordings afterwards just to obtain a larger image size: This improves the image quality Not. Software like PhotoZoom shortcut (formerly S-spline), which extrapolates the images, should therefore be treated with caution. Their results can (but do not have to) deliver an apparently better result.

transparency

Objects should, if this is possible without any loss of quality, be displayed without a background (transparent), as this makes it easier to match the color of the background of the page.

Post-processing example [edit]

Clipping [edit]

The example shows a photo of a bicycle wheel in front of a white sheet. The picture is too dark, the actually white sheet appears gray. The background (the sheet) has been cut out and replaced by a white layer. This makes the picture more informative and cleaner. It is distracted by unimportant little things like wrinkles on the sheet, and the observer only sees what matters.

The running bike was hung up in front of the sheet with a thread and was photographed from a distance of approx. 5 m with a telephoto setting. The thread was "pixelated away".

Extrapolate images [edit]

The various results are particularly clear with graphics
Less noticeable differences in a photo

Pictures that are too small can be enlarged. However, there are limits to this. Information that is not there cannot be conjured up. However, with special programs, a result can be achieved which simulates an improvement in quality to the human eye.

Real improvement only takes place in very few cases. The basic requirement for a successful extrapolation is a very good photo as the starting material. The original image must not be sharpened by camera software or an image processing program. Previous contrast changes are also critical. The following can apply as a rough rule: If an image from a 2 megapixel camera is good enough for an exposed image without any image processing (Not Photo printer) of size 20x30 cm², it can be extrapolated sometimes manage to expose it in 50x75 cm².

Extrapolation is not a good way to improve a bad picture or to blow a photo from a mobile phone into a poster!

For example on the right: The original image was 140x160 pixels. It became (from left to right, each 400%):

The Shortcut PhotoZoom (formerly S-Spline) program specializes in enlarging images, it has no other function. However, due to the quality differences to all image processing programs, it has a right to exist. The best results are achieved when one calculates with PhotoZoom to double the desired size and this result is reduced to 50% in Photoshop. Reductions and enlargements should always be made with smooth factors (1: 2, 1: 4, etc.). With good CRT monitors you can see that images in the 50% view are much sharper than in the 66% view; this effect also occurs when scaling.

Distortion of images [edit]

The rectification of photos should only be used if it is really necessary. Not every image with converging lines needs to be corrected. Equalizing, which is actually distorting, also harbors certain dangers.

In the example described, it was not possible to take a frontal picture; the focal length of the camera used (HP Photosmart M407) was not sufficient to photograph the building frontally and completely. First, the image was heavily distorted (see screenshot). The result is shown in Figure 3, the people here have become much too fat, so the entire picture was trimmed vertically again.

Reduction of transformation losses [edit]

Result comparison (please consider in 100%) with distortion: 2 = without previous scaling, 3 = with scaling.

Typical image transformations are rotations or the correction of converging lines. As a result of the transformation, various image contents are recalculated (pixel values ​​and color information). Ideally, each individual can Value before a definite one Value after be assigned.

Example: a horizontal line is shifted to the left by exactly 3 pixels. So that everyone can Value before a definite one Value after be assigned.

This ideal case practically (almost) never occurs. When recalculating the image information, not all can normally Values ​​before unambiguous Values ​​after be assigned. As a result, artifacts are created.

Example: a horizontal line is rotated 45 degrees. This means that not every old pixel value can be clearly assigned to a new value.
  • Rotating a line turns it into a "staircase", as not every old pixel value can be assigned a new pixel value. A partial compensation of this step effect is done by interpolation. In the left and right example, different types of interpolation were used.