Image Data Compression Using Windowed Weighted Channel Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
High-resolution image data from line or area scan cameras used in automated checking of printed matter results in excessive data, requiring efficient pre-selection to reduce data transmission and processing demands, while also needing to adapt to varying object surfaces in terms of color and structure.
Innovation Solution
A method and device that reduce recorded image data by specifying windows for pixel analysis, using weighted sums of intensity values from multiple channels, and prioritizing windows to focus on relevant data, allowing for flexible adaptation to different printed matter features and structures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If high-resolution line or area scan cameras are used to capture image data of printed matter, then measurement precision and detection capability are improved, but the quantity of data to be processed and transmitted increases enormously
Solution Approach 1:
The image data is divided into multiple channels (e.g., color channels and UV/infrared channels) that are processed independently. Each channel is evaluated separately to determine relevance, allowing selective retention of only those channels containing authenticating features, thus reducing overall data volume while maintaining detection precision.
Solution Approach 2:
The method extracts and retains only the relevant image data channels that contain authenticating features of the printed matter, discarding or compressing channels that do not contribute to authentication. This extraction process significantly reduces data volume while preserving measurement precision for critical features.
2Reliability
If all recorded image data channels are transmitted and processed for checking printed matter, then reliability of authentication is improved, but processing time and computational complexity increase
Solution Approach 1:
The method performs a preliminary evaluation of each image channel to determine its relevance to authenticating features before full processing occurs. This pre-screening step identifies which channels contain critical authentication information, allowing the system to focus processing resources only on relevant data, thereby reducing overall processing time while maintaining authentication reliability.
Solution Approach 2:
Different processing strategies are applied to different image channels based on their specific characteristics and relevance to authentication. Channels containing critical authenticating features receive full processing attention, while less critical channels are processed more simply or discarded, optimizing the balance between reliability and processing efficiency.
3Productivity
If a fixed data reduction method is applied to all image data, then productivity is improved through faster processing, but adaptability to different printed matter configurations decreases
Solution Approach 1:
The data reduction method dynamically adjusts which image channels are retained or discarded based on the actual content and characteristics of each printed matter item. The system adapts its processing strategy in real-time according to the detected features, allowing optimal balance between processing speed and authentication accuracy for each specific object configuration.
Solution Approach 2:
The method changes processing parameters (such as which channels to retain, compression levels, and evaluation thresholds) based on the specific characteristics of the printed matter being authenticated. This parameter adaptation allows the system to maintain high productivity across different object types while preserving adaptability to their unique configurations and features.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The invention relates to a method for reducing recorded image data from a multi-channel digital image, wherein, after detection of an object in the detection area of the sensor unit (3), the recorded digital image is output as a signal at the output of the sensor unit. According to the invention, a number of windows (11, 12, 13, 14) are specified, the row value (x) and the column value (y) of the transmitted image data are determined with respect to the digital image, it is checked whether the pixel (P) defined by the row value (x) and the column value (y) is located within a specified window (11, 12, 13, 14), and, depending on the respective window, a specified set of weights (wr, wg, wb) and a mixed value (s) are formed as a weighted sum of the intensity values or color values with the weights (wr, wg, wb).