Gyro Sensor Shake Detection with Dynamic Reference Value Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing shake detection and correction technologies in imaging devices struggle to accurately detect and correct for camera shake, especially when the device is hand-held, due to issues with drift and low-frequency noise in gyro sensor outputs, which affects the accuracy of panning detection and shake correction.
Innovation Solution
A shake detection device that includes a shake detection sensor, a storage unit for a reference value, subtraction units to remove the reference value from sensor outputs, filters to extract low and high-frequency components, and determination units to assess the device's state, allowing for accurate calculation and correction of the reference value shift, thereby enhancing shake detection and correction accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a high-pass filter is intensively applied to remove the changed reference value due to temperature drift, then the reference value accuracy is improved, but the low frequency output of the gyro sensor is removed, making it impossible to cope with slow shutter and decreasing the shake correction effect
Solution Approach 1:
The patent applies dynamics by making the cutoff frequency of the high-pass filter variable rather than fixed. The control unit dynamically adjusts the cutoff frequency based on the detected panning angular velocity: when panning is detected, the cutoff frequency is set to a first value to remove drift; when no panning is detected, it is set to a second value (lower than the first) to preserve low frequency components for slow shutter operation. This dynamic adjustment resolves the contradiction between removing drift and maintaining slow shutter capability.
2Ease of operation
If the direct current component of the sensor output before the reference value is subtracted is used for detecting panning, then the detection simplicity is improved, but it is not possible to determine whether the sensor output is drifted or a panning operation at a very low speed is performed, causing erroneous reference value detection and deteriorating shake detection accuracy
Solution Approach 1:
The patent applies preliminary action by subtracting the reference value from the sensor output before using the result for panning detection. This preliminary subtraction removes the drift component from the signal, allowing subsequent panning detection to be performed accurately without confusion between drift and actual panning motion. The control unit then detects panning based on the subtracted output, which has already had the drift removed, thus resolving the contradiction between detection simplicity and accuracy.
3Adaptability or versatility
If the reference value is updated during camera operation based on sensor output, then the adaptability to temperature changes is improved, but it is not possible to distinguish between drift and very low speed panning, causing erroneous reference value detection
Solution Approach 1:
The patent applies feedback by using the subtracted sensor output (after reference value subtraction) to detect panning motion, and then using this detected panning information to control whether the reference value should be updated. When panning is detected, the reference value update is inhibited; when no panning is detected, the reference value is updated based on the sensor output. This feedback mechanism prevents erroneous updates during panning while allowing legitimate drift compensation, resolving the contradiction between temperature adaptation and detection accuracy.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The solution enables high-accurate shake detection and correction by distinguishing between hand-held and stationary states, even at low speeds, and effectively removes noise, improving the overall precision of shake detection and correction in imaging devices.
Implementation Method 1
a gyro sensor is used for detecting shake (angular velocity) of a camera
Implementation Method 2
a low frequency component including a drift and noise is removed from the output of the gyro sensor by using a high-pass filter
Data Source
AI summary
There are provided a shake detection device of an imaging device, a shake correction device, an imaging device, and a shake detection method which are capable of performing high-accurate shake detection and shake correction. A shake detection device subtracts a reference value from a sensor output of a gyro sensor, and extracts a low frequency component and a high frequency component from a sensor output after the reference value subtraction by using an LPF and a BPF. A first determination unit determines whether or not the imaging device is in a fixed-point imaging state based on the LPF output and the BPF output. In a case where it is determined that the imaging device is in the fixed-point imaging state, a reference value shift amount calculation unit calculates a shift amount (reference value shift amount) for the reference value based on the LPF output for a period during which the determination is performed. A subtraction unit corrects the reference value by subtracting the reference value shift amount from the sensor output after the reference value subtraction. An HPF removes a low frequency noise from the sensor output after the correction of the reference value shift amount. High-accurate shake is detected.