In microscopy, pixelation is one of the most commonly occurring artifacts, as the subjects of the camera are often micro-level. Obtaining the intricate details of the bio-samples is crucial for an accurate analysis diagnosis.
Pixelation artifact occurs in imaging when the magnification surpasses an image sensor’s capabilities, leading to the appearance of indistinct pixel boundaries. It leads to a compromised image quality. This can make the interpretation of minute details of the cellular samples more tedious and error-bound.
In this blog, we are exploring pixelation artifacts, their causes, and reduction methods.
What Causes Pixelation?
Pixelation artifacts occur when the magnification level exceeds the image sensor’s capability, resulting in an invisible pixel boundary. These artifacts degrade the image quality and reduce the spatial resolution. This makes the minute details difficult to interpret. For example, the anatomical features inside a cell. As shown in the figure below.
Figure 1: Magnified Image of A Cell
Apart from resolution limitations, there are various other causes for pixelation, which are discussed below.
Digital Sampling: Digital images are created by sampling the continuous tones of the real world into discrete pixels. The spatial resolution, or pixel density, determines how finely the image can represent details. If the pixel density is insufficient to represent small details, pixelation occurs.
Zooming In or Enlarging: When an image is zoomed in or enlarged, the limited number of pixels is stretched to cover a larger display area. This stretching causes each pixel to occupy a larger portion of the image, making them more noticeable and resulting in a blocky appearance.
Lossy Compression: Compression algorithms, especially lossy compression methods like JPEG, discard some image information to reduce file size. When an image is highly compressed, details can be lost, and pixelation may occur, especially in areas with fine textures or subtle gradients.
Limited Color Depth: In some cases, pixelation can be exacerbated by a limited color palette or color depth. A lower bit depth means that there are fewer color variations available, which can contribute to a blocky appearance, especially in areas with gradual color transitions.
Overcoming Pixelation
To avoid pixelation, the camera sensors can apply Shannon’s sampling theorem.
Shannon’s Samplin Theorem: Shannon’s sampling theorem states that the camera must utilize a sampling interval that is not greater than one-half the size of the smallest resolvable feature (Figure 2).
Figure 2: Sampling Requirements of a Sensor
Here, the sampling interval is the distance between two pixels of the sensor.
To capture the smallest detail present in the subject, sampling must occur such that a minimum of two samples get collected for each feature. That is, the object of interest must be imaged onto at least two pixels of the camera sensor. But it’s also important to note that even though over-sampling provides excessive optical magnification, it provides no additional spatial information. But the fidelity of the image can be maintained by this.
What Happens If You Under sample?
Under sampling can lead to aliasing. Aliasing is the appearance of lower spatial frequency features that aren’t present in the specimen. This happens when the sampling frequency drops below a threshold value. Aliasing leads to the appearance of moire patterns. This can deter with the analysis of samples.
Figure 3: Moire Pattern
What is Nyquist Sampling Criteria?
Nyquist Sampling is, for a given microscopic configuration, the pixel size of an image needs to be at least 2.3 times smaller than the subject that is being resolved.
While imaging small features, it is important to satisfy Nyquist Sampling Criteria to avoid pixelation.
Figure 4: A Cell Image with Pixelatiom
Figure 4: A Cell Image without Pixelatiom
e-con Systems’ High-Performance Medical Cameras
e-con Systems, an industry pioneer, has 20+ years of experience in designing, developing, and manufacturing OEM cameras for various industries, including medical and life sciences.
See our newly launched e-CAM200_CUMI2020C_MOD – a 20MP high-resolution camera based on the AR2020 image sensor from onsemi’s Hyperlux LP series. This camera, with its high frame rate and compatibility with a wide range of interfaces, is an ideal option for medical imaging applications.
Visit our Medical Cameras page to see the range of medical cameras we offer. Choose the right camera from our Camera Selector Page.
If you need any help in integrating the right medical camera into your device, please write to us at camerasolutions@e-consystems.com.
Remember, we also provide customisation services.
Balaji is a camera expert with 18+ years of experience in embedded product design, camera solutions, and product development. In e-con Systems, he has built numerous camera solutions in the field of ophthalmology, laboratory equipment, dentistry, assistive technology, dermatology, and more. He has played an integral part in helping many customers build their products by integrating the right vision technology into them.