Approximate Calculation in Image Processing

(Information Science Expert) Lecture on IoT and Edge Computing

• Quite easy to slack off

• For instance, in areas of an image where it is empty, the majority of pixels tend to have similar colors

• If one can reuse the results of a calculation once done, it becomes easier to slack off (Approximation Calculation)

  • Challenge
    • How can we teach the CPU about opportunities for approximation?
      • “Pre-specifying the address where software instructions are stored”
      • Giving recently calculated/used data a sense of “freshness”
      • Saving in the Data Store Table (DST) for reuse

• Image Compression Algorithm#image Processing

  • Heavy processing areas:

• Using DST in image compression for Approximation Calculation

  • Experimental results: Choosing effective parameter combinations can reduce computational workload (compared to a regular CPU)