Understanding Image Processing

Organizations ought to have comprehensive management solutions that help them upload, save, sort, process, and retrieve their digital content in a single place. Rather than having separate systems for different file types, they'd prefer to have one system that supports all of them.

Concept

VIDIZMO offers support for all sorts of digital assets such as video, audio, image, document etc., allowing users to not just view/play them back but derive various other valuable information and insights, which is known as metadata, from within your content.

In this article, we will talk about image processing in VIDIZMO, the file formats supported for images and how to pre-configure encoding profiles to have all uploaded images processed into a set format as per requirements.

Processing of an Image

When your input image file is processed, there are certain considerations that you need to know about the output file(s) being created:

• Your original raw file is preserved which enables you to re-use the original file to create as many multiple formats and renditions for different use-cases within the teams. This is a key requirement by organizations for a comprehensive digital asset management solution to let you manage your original assets centrally, while being able to edit, make copies, save versions and distribute them for enhanced collaboration within an enterprise.
• The input image file is rendered in multiple output formats that can be custom-defined using Encoding Profiles in VIDIZMO, and this is covered later in the article.
• During processing, VIDIZMO captures information within the image (meta-data) and saves it into content attributes. These attributes are later associated with the image as custom attributes, which are then indexed against the media. These will have their own type in custom attributes called Meta-Data. You can search and access media in your portal using that meta-data such as exposure, date and time, ISO settings, etc. To learn more about search, see: How to Manage Custom Attributes and Understanding Search in VIDIZMO.
• Since media files can have a lot of attributes that may not be required for indexing or filtering such as whether the flash was on/off during capture, therefore users have the option to set which custom attributes they want to index so that they can be made searchable against the media in a portal.
• Similarly, the text is extracted from the image using OCR (Optical Character Recognition) tool and is linked to the image for enhanced searchability.
• The default target size of the image remains the same as the source image after it has been processed. For example, if someone uploads a JPEG for 1024x820 pixels, the output image will also be rendered with the same dimensions.
• While processing and creating various other renditions of the source file, additional information attached to the media is truncated to reduce file size and optimize workflow speed. This is why the data related to EXIF attributes in the input files can vary in comparison to the processed output renditions. However, these EXIF attributes are picked up and added in each of the content file renditions which can then be retrieved via the Media Info screen. To learn more, see: How to view Media Information.
• There are many underlying characteristics of an image that need to be enhanced in a manner that optimizes a viewer's experience. One such attribute is the exposure and brightness of the surroundings when the image was captured. VIDIZMO allows you to pre-configure these enhancements for output image files.

Supported File Formats

VIDIZMO employs advanced processing tools to process a huge list of input file formats and render output files after processing as per user requirements.

Here is a list of supported image formats in VIDIZMO, where Modes define whether a format is supported for only Read (R) i.e. input file format, or Write (W) i.e. output file format, or Read-Write (RW) i.e. input as well as output file format.

We support a number of other image format specifications which refer to images prepared via an algorithm, or input/output targets. The following table lists these pseudo-image formats:

Encoding Profiles

To provide greater flexibility and ease of processing, VIDIZMO processes images on the basis of which Encoding Profiles have been enabled on your portal.

Encoding profiles constitute a group of settings or configuration parameters that help determine the input/output formats of the file along with other key characteristics such as hues and depth of an image. To learn more about them, see: Understanding Encoding Profiles.

These Encoding Profiles are modifiable by the users with the added ability to create their own profiles and obtain as many renditions of an input image file as per organizational requirements.

A sample encoding profile would look like this, allowing users to define parameters like width, height, output format, image sharpening, quality, depth, etc.

{ 

"width": 1280,  

"height": 720,     

"convertTo": {   "extension": "png"    },     

"trim": false,     

"sharpen": true,     

"quality": 100,     

"depth" : 8  

} 

These parameters greatly help you custom-pick characteristics and attributes that you would want your output file to contain, which not only allows you to view your image after upload but transform it completely as per your requirements. For instance: 

• You can set "convertTo":{ "extension":"jpeg" } for retrieving an image with a jpeg extension.
• The digits you define as width and height become the pixels to size of your output image accordingly.
• Trim parameters helps you automate trimming the edges of a frame for blurred-out ends, while sharpen, if set true, visibly makes your picture clearer with more attention to detail.
• Where the value "100" is the optimum quality for an image, the standard high quality would be around 85 and the minimum lowest quality would be 0. This can vary depending upon the use of the image and the storage considerations that follow in the case of large files.
• Similarly, depth helps you define the hues of the image and their intensity. It is the amount of color information in each image pixel.  An image with a bit depth of 1 means each pixel in the image can either be black or white, as the pixel can only contain 2 values (1 bit is either "on" or "off"). As the bit depth of an image increases, the colors each pixel can contain increases. The optimal depth for an image is 8, which means 8 bits to define one pixel of an image. This gives us a huge palette of possibilities for our image, and helps enhance its color scheme.