When ordering satellite imagery, the results can sometimes be less than perfect, due to many factors. This article helps explain these factors, and why in many cases the imagery may still be considered acceptable quality.

Clouds and haze are natural elements that can appear in images depending on weather conditions. Clouds and heavy haze completely obstruct ground visibility, while light haze does not.

The following image illustrates an example of clouds and light haze. This image is considered to have approximately 50% coverage.

Best Practices:

Before selecting an image to order, review the low resolution preview image to check if any clouds or haze are covering your area of interest (AOI) layer. If clouds or haze are present inside of your AOI, then the delivered product will contain them.

It is most recommended to select a product that contains no clouds or haze.

You can filter returned products by Cloud Cover using the filter tool in search.

Example #1: Preview Image Containing Clouds in AOI

Example #2: Preview Image Containing No Clouds in AOI

Image noise refers to the random variation in brightness or color information in images, and is also known as graininess, speckling, or salt and pepper.

Noise is common in low-light conditions, where the image is enhanced to reveal features that are otherwise not visible, which can introduce noise.

Noise is often found in images taken during cloudy, hazy, or other low-light conditions for these reasons.

Unless the noise makes the features in the image unrecognizable, it should not be considered a defect.

The following image shows noise resulting from a combination of haze and low lighting conditions.

Stitching of images, also known as mosaicking, involves joining multiple images into a single, cohesive image.

Since the images being mosaicked are taken at different times, they may have different lighting conditions, colours (e.g., grass appearing more green or brown), and cloud cover (e.g., light haze causing noise). As a result, a seam or stitch line may be visible where the images meet, due to these differences. Although SkyWatch employs various image processing techniques to blend stitch lines, these methods typically reduce but do not completely eliminate them.

Additionally, the georeferencing of each image (how accurately it is located relative to the Earth) may differ slightly. When the images are joined, this may result in a small discontinuity.

A stitch line is not considered a defect as long as the georeferencing between the two images is within 20 meters.

The following mosaicked image uses two images taken one day apart. Due to natural variations in lighting and tide, the contrast between the water level and lighting is noticeable along the stitch line, but this is not considered a defect.

Missing parts of an image can occur when there is insufficient satellite coverage available.

In some cases, the satellite data from the search results may only cover a portion of the Area of Interest (AOI), leaving the uncovered areas missing or blank.

Example:

Chromatic aberration occurs when the red, blue, and green bands misalign in certain areas of an image due to natural phenomena during the satellite camera’s capture process. Since this misalignment is a natural result of the satellite's observation, it should not be considered an image defect.

Refraction from cloud/haze:

Intense reflection from a rooftop on a sunny day:

Moving object, such as a car or airplane: