Interpolation happens when a camera’s native sensor size is 5MP (megapixel) for example and the owner sets the camera in a 12MP setting on the menu options.
This is a digital process of reinventing a picture by filling in the holes. It means the camera firmware is stretching the image to a larger size and thus must use a mathematical process to compare neighboring pixels and interpolate a new pixel based on an average color scheme. When the picture is sampled and viewed without enlargement all the pixels are laying side by side. As we enlarge or increase the size, holes begin to develop and there is a need to fill in that space with a color that is a combination of the neighboring pixels.
This can be readily seen when you attempt to over enlarge a picture and the little jagged edges and pixel squares start to become apparent. To produce an image that is larger than what the cameras sensor captured or digitally zoom that image you must fill in the space with pixels that are an average color of the surrounding pixels. This produces a more blended effect but not a sharper effect because this process cannot clean or sharpen the picture better than the original image in its native size.
There are several methods of interpolation used in the camera industry. Bicubic being the most sophisticated followed by bilinear and the one discussed above. Our trail cameras probably only use the first discussed method where near pixels are sampled for color and reintroduced next door.
I have never been a fan of this type of digital enhancement in our game cameras and feel that anything over 3 to 5 MP is over kill. If the manufacturers used a high quality CMOS sensor, the need for all of this digital enhancement would go away. By taking an image from your camera and then using one of the photo enhancement programs on your computer, you are using some method of interpolation in order to achieve the desired effect. I believe that it is obvious that it is cheaper for the camera companies to electronically interpolate rather than to provide quality image sensors in the first place.