Line Chart Data Filtering for Rendering Performance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing path simplification algorithms for line and area charts smooth out local extrema and skip points unevenly, leading to data misrepresentation and slow rendering performance, especially when dealing with uneven data point intervals.
Innovation Solution
The proposed solution filters data by ensuring all visible local maxima and minima are rendered without alteration and skipping points evenly, using a two-point-per-pixel approach to maintain visual uniformity and achieve O(n) algorithmic complexity, significantly improving chart performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If path simplification algorithms are used to reduce the number of points drawn along the line, then rendering performance is improved, but local extrema are smoothed out and data misrepresentation occurs
Solution Approach 1:
The patent segments the data points into three categories: local extrema points (maxima and minima), intermediate points, and endpoint. By treating these segments differently - preserving extrema points while simplifying intermediate points - the solution maintains data accuracy at critical locations while achieving rendering performance improvement through selective point reduction.
Solution Approach 2:
The patent applies different quality requirements to different parts of the data visualization. Local extrema points are preserved with high fidelity to maintain data accuracy, while intermediate points are simplified more aggressively. This local differentiation allows the system to achieve overall rendering performance improvement without sacrificing critical data representation.
2Productivity
If existing path simplification algorithms are used, then the number of drawn points is reduced, but points are skipped unevenly causing visual inconsistency
Solution Approach 1:
The patent implements a dynamic point skipping strategy where the number of points to skip varies based on the local characteristics of the data. The algorithm calculates an optimal skip count for each segment based on the distance between local extrema and the desired visual density, creating a dynamic adaptation to different data regions rather than applying a uniform simplification rule.
Solution Approach 2:
The patent changes the parameter of point density distribution from uniform to non-uniform based on data characteristics. By adjusting the skip count parameter dynamically according to the distance between extrema and the desired visual uniformity, the system achieves both rendering speed improvement and visual consistency across different regions of the chart.
3Loss of information
If all data points are drawn, then data accuracy is maintained, but rendering becomes unbearably slow for large datasets
Solution Approach 1:
The patent extracts and preserves only the critical data points (local extrema) that carry the most important information for data interpretation. By taking out these essential points and selectively simplifying or removing less critical intermediate points, the system maintains data completeness in terms of important features while achieving significant rendering speed improvement through reduced point count.
Data Source
AI summary
Systems and methods are disclosed that, in various embodiments, improve chart performance by data filtering. In one aspect, the disclosed techniques allow all visible local maxima and minima to be rendered without having their values altered. In another aspect, the disclosed techniques filter data by skipping points evenly. In some embodiments, every horizontal pixel contains two points, so the line is visually uniform. This can dramatically enhance visualizations where data point intervals are uneven (e.g., having “mixed frequency”). Using a two point per pixel approach, the disclosed techniques ensure that a resulting simplified line looks almost identical to the original line by drawing at the limit of the screen resolution. The disclosed techniques allow for an algorithmic complexity of O(n), which is significantly faster than existing solutions.