71 results about "Fixed-pattern noise" patented technology

The invention relates to the field of computational imaging, in particular to a computational imaging method of a crossed strip-shaped anode single-photon detector, which comprises the following steps of: screening effective data through full-view-field illumination, and calculating an average output signal amplitude of each anode strip and a total average output signal amplitude of all anode strips to obtain a correction factor; the method comprises the following steps of: correcting an original output signal of each anode strip by adopting a correction factor during target imaging, performing effective signal data screening on the corrected original output signal, substituting effective signal data into a Gaussian algorithm to calculate an incidence position coordinate, mapping the incidence position coordinate to a mapping matrix, and stretching the incidence position coordinate to a 256 gray level for display. According to the calculation imaging method provided by the invention, the fixed mode noise can be suppressed, the imaging resolution is improved, and the image distortion is reduced.

A method for detecting linear fixed pattern noise (FPN) comprises extracting a plurality of pixel values from a first row of the ROI, calculating a brightness value of the first row based on the extracted pixel values, calculating a first score indicating the variance of the pixel values of the first row, calculating a second score indicating a difference between the pixel value average of the first row and the pixel value average of the adjacent row to the first row, and determining whether the fixed pattern noise is present in the first row based on the first score and the first threshold and the second score and the second threshold.

This disclosure is directed to an image authenticity detection method and apparatus. The method includes: obtaining an image; removing low-frequency information from the image to obtain first image information of the image; denoising the first image information to obtain second image information; determining, based on a difference between the first image information and the second image information, a fixed pattern noise feature map corresponding to the image; analyzing distribution of fixed pattern noise in the fixed pattern noise feature map, the fixed pattern noise being inherent noise from a camera sensor and not interfered by image content; and detecting, based on the distribution, authenticity of the image to obtain an authenticity detection result of the image.

A method for detecting linear fixed pattern noise (FPN) includes: extracting a plurality of pixel values from a first row of an ROI; calculating a luminance value of the first row based on the extracted pixel values; calculating a first score indicative of a variance of pixel values of the first row; calculating a second score indicating a difference between an average value of pixel values of the first row and an average value of pixel values of adjacent rows of the first row; and determining whether there is fixed pattern noise in the first row based on the first score and a first threshold and the second score and a second threshold.

The disclosure extends to methods, systems, and computer program products for producing an image in light deficient environments having cancelled fixed pattern noise.

Provided are an image processing device configured to correct fine fixed pattern noise (FPN), an image processing system, and an operating method thereof. Provided is an operating method of an image processing device, the method including receiving at least one image group including a plurality of images, detecting, from the at least one image group, fixed pattern noise (FPN) information using Fast Fourier Transform (FFT), and correcting FPN based on the FPN information.

A method for correcting fixed-pattern noise in at least one image is disclosed. The method includes the following steps:—acquiring at least N substantially distinct and regular images by means of at least one rotating image sensor such that the combination of the N images forms a panorama of a surrounding scene, the fixed-pattern noise in each of the N acquired images being substantially the same;—determining at least one correction parameter for correcting the fixed-pattern noise by means of a processing module from the at least N acquired images, the at least one correction parameter minimizing a functional;—correcting the fixed-pattern noise in each of the N images acquired by the processing module using the at least one determined fixed-pattern noise correction parameter.

Disclosed herein are fixed pattern noise (FPN) reduction techniques in image sensors operated with pulsed illumination. In one embodiment, a method includes, during a first sub-exposure period of a frame, (a) operating a first tap of a pixel to capture a first signal corresponding to a first charge at a first floating diffusion, the first charge corresponding to a first light incident on a photosensor, and (b) operating a second tap of the pixel to capture a first parasitic signal corresponding to FPN at a second floating diffusion. The method further includes, during a second sub-exposure period of the frame, (a) operating the second tap to capture a second signal corresponding to a second charge at the second floating diffusion, the second charge corresponding to a second light incident on the photosensor, and (b) operating the first tap to capture a second parasitic signal corresponding to FPN at the first floating diffusion.

The invention provides an infrared image non-uniformity correction method, image processing equipment and a storage medium, and the method comprises the steps: subtracting estimated global fixed pattern noise from a current observation frame, obtaining a preliminary correction image, carrying out the analysis and calculation of local fixed pattern noise based on continuous multiple frames of preliminary correction images, and carrying out the secondary filtering of the local fixed pattern noise. According to the two-stage processing mechanism, the observation image is corrected based on the inherent global fixed pattern noise of the infrared focal plane array and the local fixed pattern noise generated due to scene change, time consistency and space difference are considered, and the fixed pattern noise is effectively removed. According to the method, extra calibration is not needed, the fixed pattern noise can be stably extracted, the dynamic self-adaptive capability is achieved, and the quality and usability of the infrared image in a complex environment are improved.

The invention relates to the technical field of image processing, and discloses a pixel signal sampling circuit, a sampling method, an image sensor and electronic equipment, the pixel signal sampling circuit comprises a first sampling branch, a second sampling branch and a reset branch, the first sampling branch is used for collecting each pixel reset signal to a control module, and the second sampling branch is used for collecting each pixel reset signal; the second sampling branch is used for collecting exposure signals of all pixels to the control module, the two ends of the reset branch are electrically connected with the output end of the first sampling branch and the output end of the second sampling branch respectively, and the reset branch is used for being connected after the first sampling branch and the second sampling branch output respective signals each time. And the output ends of the first sampling branch and the second sampling branch are reset. The reset branch circuit is conducted after the pixel exposure signal and the pixel reset signal of the pixel are collected each time, reset is carried out, the difference between the initial states of different pixel units is eliminated, therefore, fixed mode noise is reduced, and the quality and consistency of images are greatly improved.

Method for detecting faulty pixel addressing in an image sensor comprising a pixel array, wherein the method comprises: Obtaining an actual fixed pattern noise for targeted pixel addresses of the image sensor; Correlate the actual fixed pattern noise with a reference fixed pattern noise known for the targeted pixel addresses; and Detecting an addressing error for the image sensor based on the correlation results.

This application provides a signal processing device, image sensor, electronic device, and signal processing method, relating to the field of sensor technology. It improves column fixed-pattern noise in image sensors and enhances image quality. Specifically, the signal processing device includes a switching module, a switching decoder, a multiplexer, and multiple signal conversion modules. Multiple input terminals of the switching module are coupled one-to-one with multiple pixel circuits; multiple output terminals of the switching module are coupled one-to-one with the input terminals of the multiple signal conversion modules; the control terminal of the switching module is coupled with the output terminal of the switching decoder; and the input terminal of the switching decoder is coupled with a timing controller. The output terminals of the multiple signal conversion modules are coupled one-to-one with the multiple input terminals of the multiplexer; the output terminal of the multiplexer is used to output the quantized pixel signal. This application embodiment describes the process of reading out the pixel signal.

The invention discloses a photovoltaic cell photoluminescence image noise identification and suppression method and system, and relates to the technical field of photovoltaic cell defect detection.The method comprises the steps that photoluminescence images of a photovoltaic cell under dark field and bright field conditions are collected through an image sensor; pre-processing the image, wherein the pre-processing comprises blind pixel correction based on a manual calibration method and black level correction based on absolute difference operation; based on the preprocessed image, fixed pattern noise is identified through a multi-frame comparison technology, and a noise position mapping matrix is generated; and based on the matrix, carrying out noise suppression on the image by adopting a selective multi-stage median filtering strategy, and outputting a corrected image. According to the method, the problems of image quality reduction and low defect identification precision caused by fixed-mode noise interference in the prior art are effectively solved, efficient processing and detail reservation are taken into account, and the accuracy and reliability of photovoltaic cell defect detection are remarkably improved.

A nuclear-grade radiation-resistant infrared video monitoring system belongs to the technical field of nuclear facility safety monitoring, and comprises a radiation-resistant infrared camera used for collecting infrared video data in a strong nuclear radiation environment; the image processing unit is used for performing enhancement processing on the infrared video data, removing non-uniform fixed mode noise caused by radiation and obtaining an enhanced video image; and the intelligent analysis and prediction unit is used for carrying out quantitative identification on the enhanced video image, obtaining pipeline defects in the video image and carrying out risk prediction. The problems that no available infrared monitoring equipment exists in a radiation area of an existing nuclear facility, the existing nuclear facility depends on an imported visible light camera (the problems of short service life, sensitivity to temperature, poor image quality and the like), and the existing nuclear facility lacks intelligent early warning capability are solved.

Thermal imaging systems can include an infrared camera module, a user interface, a processor, and a memory. The memory can include instructions to cause the processor to perform a method upon a detected actuation from the user interface. The method can include performing a non-uniformity correction to reduce or eliminate fixed pattern noise from infrared image data from the infrared camera module. The method can include capturing a infrared images at a plurality of times and register the captured images via a stabilization process. The registered, non-uniformity corrected images can be used to perform a gas imaging process. A processor can be configured to compare an apparent background temperature in each of a plurality of regions of infrared image data to a target gas temperature. The processor can determine if such regions lack sufficient contrast to reliably observe the target gas.

A radiation imaging apparatus including pixels each including a conversion element and a switch element, a signal line to which signals are supplied from the pixels, a readout circuit reading out a signal supplied to the signal line and a processor processing a signal read out by the readout circuit, is provided. The readout circuit performs, during irradiation with radiation, a first operation of reading out a signal while the switch element of each of the pixels is in a non-conduction state and a second operation of reading out a signal while the switch element of a predetermined pixel of the pixels is in a conduction state. The processor corrects a second signal value of a signal read out by the second operation by using a first signal value of a signal read out by the first operation.

The invention discloses a fixed pattern noise detection method for a camera module. The method comprises the following steps: step 100, obtaining a test image of the camera module; step 200, decomposing the test image into a plurality of single-channel images according to the number of channels of the test image; step 300, calculating single-row pixel values and single-column pixel values of all the single-channel images; and step 400, determining whether the test image has fixed pattern noise according to the single-row pixel values and the single-column pixel values of all the single-channel images. The fixed pattern noise detection method can improve the detection and identification precision of the fixed pattern noise. The invention further discloses a fixed pattern noise detection device for the camera module, which is used for realizing the method.

Techniques are provided for removing fixed pattern noise in image frames using noise estimate values and flatness measurement values. In one example, a method includes selecting an image frame comprising a plurality of pixels. The method also includes calculating, for each pixel, a noise estimate value and a flatness measurement value. The method also includes determining, for each pixel, a damping factor using the flatness measurement value. The method also includes selectively weighting by the damping factor, for each pixel, the noise estimate value associated with the selected image frame and a previous noise estimate value associated with a previous image frame to generate a noise correction term. Additional methods and systems are also provided.

Sensing element transducers in a sensing system are connected to sensing element electronics by a switchable averaging network. The switchable averaging network may be configured to cause the sensing elements to operate in a sensing mode where the sensing elements all operate independently or in a calibration mode where the connected sensing elements have as an input a uniform signal generated by combining signals from all of the interconnected sensing element transducers. Signals output from the sensing system in the calibration mode may differ from the uniform signal due to unstable FPN (fixed-pattern noise) generated by the sensing element electronics and readout channels. These differences may be derived by comparing the signals output from the sensing system in the calibration mode to the uniform signal, and may be used to reduce FPN.

A method of an optical navigation device to be coupled to a host device includes: providing a pixel array circuit to capture and generate an image frame; performing a bandpass filter operation upon the image frame to remove a low-frequency image noise and a high-frequency image noise from the image frame so as to generate a bandpass-filtered image frame; and performing a correlation operation upon the bandpass-filtered image frame and a reference image frame to generate a displacement result between the bandpass-filtered image frame and the reference image frame as a motion result of the optical navigation device; the displacement result is to be reported to the host device if the displacement result indicates a motion.

Fixed pattern noise (FPN) reduction techniques in image sensors operated with pulse illumination are disclosed herein. In one embodiment, a method includes, during a first sub-exposure period of a frame, (a) operating a first tap of a pixel to capture a first signal corresponding to first charge at a first floating diffusion, the first charge corresponding to first light incident on a photosensor, and (b) operating a second tap of the pixel to capture a first parasitic signal corresponding to FPN at a second floating diffusion. The method further includes, during a second sub-exposure period of the frame, (a) operating the second tap to capture a second signal corresponding to second charge at the second floating diffusion, the second charge corresponding to second light incident on the photosensor, and (b) operating the first tap to capture a second parasitic signal corresponding to FPN at the first floating diffusion.

This invention provides a fixed-pattern noise reduction method based on an encoder-decoder network, comprising: S1. Acquiring RAW images using an electronic imaging device equipped with a target image sensor, and creating a paired dataset for training a denoising model; S2. Performing necessary preprocessing on the dataset, expanding the dataset size and increasing its diversity through data augmentation; S3. Designing a lightweight encoder-decoder network, which consists of an encoder, connection layers, a decoder, and an output layer; S4. Training the encoder-decoder network based on the preprocessed paired dataset; S5. After the model training reaches convergence, loading the preprocessed, uncropped complete image into the model, and obtaining the denoised image through post-processing. This method utilizes the powerful fitting ability of the encoder-decoder network to eliminate prominent fixed-pattern noise in low-light images, and is applicable to different types of image sensors; the designed encoder-decoder network has low computational cost, making it suitable for resource-constrained edge devices.

The application discloses a two-stage enhanced camera response calibration method, which comprises the following steps: fixed pattern noise filtering on a to-be-calibrated image; collecting response images of an enhanced camera under continuous linear enhanced input light intensity conditions; calculating the change relationship between the difference between the actual output response of a pixel and the standard response of the pixel under different input light intensity conditions and the input light intensity based on the response images under different input light intensity conditions; performing primary response error compensation on the to-be-calibrated image after noise filtering based on the change relationship between the difference between the actual output response of a pixel and the standard response of the pixel under different input light intensity conditions and the input light intensity; and performing secondary response error compensation on the to-be-calibrated image after the primary response error compensation based on a pre-trained machine learning algorithm to obtain the final calibration result corresponding to the to-be-calibrated image. The application can significantly improve the sensitivity, accuracy and linearity of the response of an enhanced imaging system to an input signal.

An imaging device includes: an image sensing device, which includes one or more first image detection pixels and one or more second image detection pixels configured to convert light incident on the image detection pixels into pixel signals representing an image; a first transistor, which is connected to the one or more first image detection pixels and is configured to generate physically unclonable function PUF data including information about at least one fixed pattern noise FPN data and at least one random telegraph noise RTN data; and a second transistor, which is connected to the one or more second image detection pixels and is configured to generate image data acquired by converting light corresponding to the image; a private key generation unit, which is configured to generate a private key based on at least one FPN data and at least one RTN data of the PUF data; and an image encryption unit, which is configured to encrypt the image data using the private key.

The invention discloses a low-noise readout circuit for a medium-wave infrared detector and a control method of the low-noise readout circuit. The low-noise readout circuit comprises a pixel-level circuit module, a sampling and amplifying circuit module and a monoclinic column-level ADC (Analog to Digital Converter) module, the pixel-level circuit module is used for sampling photoelectric signals generated by the infrared detector; the sampling and amplifying circuit module is used for sampling and amplifying the pixel reset signal and the photoelectric signal output by the pixel-level circuit module; the monoclinic column-level ADC module is used for quantizing the analog voltage values output by the sampling amplification circuit module and outputting the difference value of two-time quantized digital codes; wherein each infrared detector in the full array corresponds to one pixel-level circuit module, and each column of infrared detectors in the full array corresponds to one sampling amplification circuit module and one monoclinic column-level ADC module. According to the low-noise reading circuit, fixed-mode noise and low-frequency noise in the low-noise reading circuit of the medium-wave infrared detector can be effectively suppressed, and the noise level of the whole reading circuit is reduced.

The invention relates to an image sensor fixed pattern noise calibration system and method based on an FPGA, and the system comprises a parameter storage module which is used for storing calibrated row FPN calibration parameters and column FPN calibration parameters, the row FPN calibration parameters are stored according to the number of rows of an image sensor, and the column FPN calibration parameters are stored according to the number of columns of the image sensor; the column FPN calibration parameters are stored according to the number of columns of the image sensor. And the line calibration module is used for receiving the image data of the image sensor and carrying out calibration based on the line FPN calibration parameters to obtain line calibration result parameters. And the column calibration module is connected with the row calibration module, and is used for receiving the row calibration result parameter, performing calibration based on the row calibration result parameter and the column FPN calibration parameter, and obtaining the calibrated image data in the multi-slope HDR mode. The problem of how to effectively correct the multi-slope HDR mode fixed pattern noise under the constraint of limited storage resources in the prior art is solved.

An imaging apparatus (201) is provided in which a sidewall spacer (74) width of a transfer gate device (TX) is greater than a sidewall spacer (71, 72, 73) width of other in-pixel devices. According to the invention, GIDL and GISL in a photodiode (PD) region and a floating diffusion (FD) region can be reduced, and contact and grid spacing of devices in pixels can be reduced at the same time.

The invention relates to an imaging system and a method of operating an image sensor. The imaging system may include an image sensor having an array of pixels arranged in rows and columns. A row driver circuit may be coupled to the pixel and may address a given row of the pixel based on a row address. Row randomization circuitry may be coupled to the row driver circuitry to dither the row address and thus randomize the order of reading out rows of a given column through column lines to sample and hold circuitry, the sample and hold circuitry including a plurality of capacitors for each column. The imaging system may include a multiplexer that de-randomizes the order of the row after passing through the sample and hold circuit. Alternatively or additionally, the order in which the signals from each column are stored using the plurality of capacitors may be randomized. In this manner, fixed pattern noise in the output image may be reduced or eliminated.

According to the circuit structure for the X-ray detector, the circuit structure is formed on the substrate by adopting the ink through a printing process. The circuit structure comprises a row driving circuit and a column acquisition circuit, wherein each of the row driving circuit and the column acquisition circuit comprises a gate line, a gate insulating layer, a semiconductor layer, a source line, a drain line and a protective layer. The gate insulating layer covers the gate line. The semiconductor layer covers the gate insulating layer. The source line and the drain line are arranged on the upper surface of the semiconductor layer. The protection layer covers the source line and the drain line. The circuit structure is formed by printing ink through a printing technology, in the X-ray imaging process, the circuit structure hardly generates absorptive'development ', fixed-mode noise such as image stripes and shadows caused by internal metal wiring is effectively avoided, and therefore the signal-to-noise ratio and uniformity of imaging are improved.

Dynamic systems and methods for fixed noise pattern calibrations and image corrections are provided. Imaging devices may be calibrated via the system by adjusting an imaging sensor exposure to acquire one or more black images. A global average pixel value is calculated to accumulate pixel values per column of the acquired image. The calculated global average pixel value is then saved as offset imaging data. An average pixel value per column of the imaging data based on the accumulated pixel values may also be calculated, and stored as offset imaging data. A correction matrix may be recomposed based on the offset imaging data and used for calibrating the imaging sensor, and for creating corrected images in or near real time.