Closed-Loop ISP Throttling for Flat-Pixel Power Control

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current image signal processors (ISPs) face high power consumption issues due to their high frames-per-second and resolution imaging capabilities, leading to increased noise and processing loads, which degrade sensor performance and image quality, as existing methods operate in an open-loop mode and cannot effectively measure or control processing loads.

Innovation Solution

A closed-loop dynamic tuning method is employed to control power consumption in ISPs by measuring the actual percentage of flat pixels in digital images, adjusting a percentage threshold based on measured values using a proportional-integral-derivative (PID) controller, and processing images accordingly to reduce power usage, potentially disabling complex processing tasks like direction and corner detection.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If high frames-per-second and resolution imaging is performed, then image quality and processing capability are improved, but power consumption increases

Engineering Contradiction:
Improveframes-per-secondVSAvoidpower consumption
Core Design Contradiction:
ProductivityVSUse of energy by moving object

Solution Approach 1:

The patent applies dynamic tuning of the flat pixel percentage threshold based on real-time measurement of actual flat pixel percentages in video sequences. This allows the ISP to adaptively adjust processing intensity according to scene characteristics, enabling high frame rates when possible while reducing power consumption when scenes allow for simpler processing.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent changes the operational parameters of the ISP by dynamically adjusting the flat pixel percentage threshold. This parameter adjustment enables the system to modulate processing load and power consumption while maintaining acceptable image quality, directly addressing the contradiction between high productivity and energy efficiency.

Inventive Principle:
Principle #35Parameter changes

2Measurement precision

If detailed analysis of image patches is performed, then processing accuracy is improved, but power consumption increases

Engineering Contradiction:
Improveprocessing accuracyVSAvoidpower consumption
Core Design Contradiction:
Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The patent applies local quality by differentiating processing intensity based on local image characteristics. By measuring the actual percentage of flat pixels in different regions and time periods, the system applies detailed analysis only where necessary while using simpler processing for uniform regions, thereby reducing overall power consumption while maintaining processing accuracy where needed.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The patent implements partial action by selectively applying detailed image patch analysis only when the measured flat pixel percentage indicates it is necessary. For scenes with high flat pixel percentages, the system reduces processing intensity, performing partial analysis rather than full detailed analysis, thus reducing power consumption while maintaining adequate processing accuracy.

Inventive Principle:
Principle #16Partial or excessive action

3Object-affected harmful factors

If complex processing tasks are performed to overcome noise, then image quality is improved, but power consumption increases

Engineering Contradiction:
Improvenoise levelVSAvoidpower consumption
Core Design Contradiction:
Object-affected harmful factorsVSUse of energy by moving object

Solution Approach 1:

The patent implements feedback control by continuously measuring the actual percentage of flat pixels in video sequences and using this information to adjust the flat pixel percentage threshold for subsequent processing. This closed-loop approach allows the system to respond to changing scene conditions and noise levels, adjusting processing intensity to overcome noise only when necessary, thereby managing power consumption effectively.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The patent applies periodic action by measuring flat pixel percentages over multiple frames and time periods to determine appropriate processing intensity. This periodic measurement and adjustment approach allows the system to account for temporal variations in scene content and noise levels, applying complex processing tasks periodically rather than continuously, thus reducing overall power consumption.

Inventive Principle:
Principle #19Periodic action

4Device complexity

If open-loop processing settings are applied, then device complexity is reduced, but power consumption control effectiveness deteriorates

Engineering Contradiction:
Improveprocessing controlVSAvoidpower consumption control
Core Design Contradiction:
Device complexityVSUse of energy by moving object

Solution Approach 1:

The patent transitions from open-loop to closed-loop control by implementing feedback based on actual measurement of flat pixel percentages. The system measures the actual percentage in each video sequence and adjusts the threshold accordingly, providing effective power consumption control without significantly increasing device complexity. This feedback mechanism enables adaptive power management based on real scene conditions.

Inventive Principle:
Principle #23Feedback

Data Source

PatentUS11812139B1Closed loop dynamic ISP throttling
Publication Date: 2023.11.07 SAMSUNG ELECTRONICS CO LTD
  • US11812139B1 patent drawing
  • US11812139B1 patent drawing
  • US11812139B1 patent drawing

AI summary

A method of controlling power usage in an image signal processor includes measuring an actual percentage of flat pixels in a digital image, adjusting a percentage threshold of flat pixels in the digital image, based on the measured actual percentage of flat pixels in the digital image; and processing the digital image by the image signal processor based on the percentage threshold of flat pixels. The percentage threshold of flat pixels is adjusted to reduce power used by the image signal processor in processing the digital image.