Method for correcting defects, in particular for reducing noise in an image provided by an image sensor - Patents.com

Abstract

To correct defects that appear in an image in consideration of the temperature of an image sensor and an integration time.SOLUTION: An image to be corrected taken by an image sensor, the temperature of the image sensor acquired when the image to be corrected is taken, and an applied integration time are received. A defect is corrected by subtracting a pixel-specific noise correction factor from the pixel value of each pixel px[i, j]. The noise correction factor bm[i, j] is determined by the formula: bm[i, j]=IT×ad[i, j]×Exp(bd[i, j]×TP)+ab[i, j]×TP+bb[i, j], IT denotes the integration time, TP the temperature of the image sensor, EXP an exponential function, and ad[i, j], bd[i, j], ab[i, j], and bb[i, j] denote pixel-specific coefficients.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to the field of imaging devices such as video and still cameras. The invention particularly relates to imaging devices incorporating image sensors in CMOS (Complementary Metal Oxide Semiconductor) or CCD (Charge Coupled Device) technology of the CTIA ("Charge Transimpedance Amplifier") or SF ("Source Follower") type. The invention can be applied to the field of imaging in the visible range as well as in the SWIR, MWIR and LWIR ("Short, Mid and Long Wave Infrared") regions. [Background technology]

[0002] Typically, a CMOS image sensor consists of pixels, or photosites, arranged in an array configuration. Each pixel comprises a light-sensitive region (usually a photodiode) configured to accumulate charge according to the light it receives, and a readout circuit for measuring the amount of charge accumulated by the photodiode. The readout circuit comprises a transfer transistor for controlling the transfer of the charge accumulated in the photodiode to a readout node. The pixel is thus controlled according to a cycle including an initialization phase, an integration phase, and a readout phase. During the integration phase, the photodiode accumulates charge according to the light it receives. The readout phase involves generating a signal corresponding to the amount of charge accumulated by the photodiode during the integration phase. The initialization phase involves removing the charge accumulated by the photodiode during the integration phase.

[0003] Images produced by modern image sensors are disturbed by various noise sources, especially as image sensors are miniaturized. The effects of some of these noise sources are evident in images captured when the image sensor is placed in a dark environment. Under these conditions, the pixel circuits of the image sensor may generate a signal, but the pixels are completely dark. The amplitude of these dark signals has been found to vary with the temperature of the image sensor and the integration time selected to produce the image. The amplitude of these dark signals varies from pixel circuit to pixel circuit within the same image sensor, even those from the same manufacturing batch, and also from image sensor to image sensor.

[0004] It is known to establish a fixed pattern noise image by generating a dark image using an image sensor placed in a dark location and subtracting this dark image from the image generated by the image sensor. To account for temperature variations and integration time, it would be necessary to generate such a dark image each time a new image is acquired. However, in many applications, particularly video capture, the time available before each frame is often insufficient to generate such a dark image. Furthermore, in the case of video images, applying corrections to each frame or series of frames inevitably introduces flicker that can disturb them. Summary of the Invention

[0005] It is therefore desirable to provide an efficient method for noise reduction in imaging devices that takes into account variations in image sensor temperature and changes in integration time, and that does not disrupt the viewing experience, and that is applicable to each imaging device.

[0006] Embodiments relate to a method of correcting defects appearing in an image produced by an image sensor, the method including receiving an image to be corrected captured by the image sensor; receiving a temperature from the image sensor acquired when the image to be corrected was captured by the image sensor; receiving an integration time applied by the image sensor for each pixel of the image to be corrected when capturing the image to be corrected; and subtracting from pixel values ​​a pixel-specific noise correction factor derived from a noise reduction model including a linear component dependent on the temperature of the image sensor added to an exponential component dependent on the temperature of the image sensor and multiplied by the integration time, the linear and exponential components being determined by pixel-specific coefficients.

[0007] According to one embodiment, the noise reduction model is represented by the following equation: bm[i,j]=IT×ad[i,j]×Exp(bd[i,j]×TP)+ab[i,j]×TP+bb[i,j] where bm[i,j] is the noise correction factor to be subtracted from the corresponding pixel of the image to be corrected, IT is the integration time, TP is the temperature of the image sensor, EXP is an exponential function, and ad[i,j], bd[i,j], ab[i,j] and bb[i,j] are pixel-specific coefficients. It is determined by:

[0008] According to one embodiment, the method includes calculating a noise correction factor for each pixel each time the integration time is changed or the temperature of the image sensor deviates from a previous value by more than a temperature deviation threshold.

[0009] According to one embodiment, the method includes acquiring images with an image sensor in the absence of light at a minimum integration time, each image being taken at a different respective temperature, and determining coefficients of linear components for each pixel of the image to be corrected by a linear regression calculation applied to corresponding pixels in the images taken at each temperature in the absence of light.

[0010] According to one embodiment, the method includes acquiring images by an image sensor in the absence of light at different integration times, exposing the image sensor to different temperatures, generating corrected images by subtracting an image taken by the image sensor in the absence of light at a minimum integration time and the same temperature from each of the images taken by the image sensor in the absence of light at different integration times and different temperatures, and determining coefficients of exponential components by an exponential fitting calculation applied to each pixel of the corrected images corresponding to the same integration time obtained at the different temperatures.

[0011] According to one embodiment, the coefficients of the exponential component are determined by averaging the coefficients obtained by the exponential fitting calculation for different integration times.

[0012] According to one embodiment, the noise reduction model includes components that are identical for all pixels of the image sensor depending on the integration time.

[0013] According to one embodiment, the method includes a step of obtaining a video stream image, wherein a noise correction factor corresponding to each pixel of the image of the video stream is subtracted from the corresponding pixel of each image of the video stream.

[0014] According to one embodiment, the method includes receiving a command to select a gain value for the image sensor; and selecting a pixel-specific noise correction factor for each pixel of the image to be corrected as a function of the selected gain value, the noise correction factor being used to correct a value of each pixel of the image to be corrected, the noise correction factor being determined from a set of pixel-specific coefficients generated as a function of the selected gain value.

[0015] According to one embodiment, the method includes, for each pixel of the corrected image after noise reduction, multiplying the pixel's value by a pixel-specific gain correction factor selected from a gain normalization table to obtain an image with normalized gain.

[0016] According to one embodiment, the method includes updating the gain normalization table each time the integration time is changed or the temperature of the image sensor deviates from a previous value by more than a temperature deviation threshold, where updating the gain normalization table is performed by an interpolation calculation applied to a set of gain normalization tables determined for different temperatures.

[0017] According to one embodiment, the method includes acquiring a series of images with an image sensor in the presence of a uniform light source at different integration times or different light source intensities, each series of images being taken at a different respective temperature; determining, for each pixel of an image in each series of images, a gain by a linear regression calculation applied to a corresponding pixel in the image in the series of images; and determining, for each pixel of an image in each series of images, a gain correction factor by dividing the average of gains obtained for all pixels in the series of images by the gain determined for the pixel.

[0018] According to one embodiment, the method includes acquiring a series of images by an image sensor in the absence of light at an average integration time, each image being obtained by the image sensor exposed to a respective temperature; calculating an average deviation at different image sensor temperatures for each pixel of one of the acquired series of images, each deviation being calculated for one image sensor temperature between the value of that pixel of the series of images corresponding to the image sensor temperature and a noise correction factor defined for that pixel at the average integration time and image sensor temperature; comparing the average deviation to a threshold value and considering the pixel to be defective if the average deviation is greater than the threshold value for the pixel.

[0019] According to one embodiment, the method includes correcting each image acquired by the image sensor by replacing the value of a defective pixel with the value of an adjacent pixel or an average value of adjacent pixels, or correcting each image acquired by the image sensor by replacing the value of a defective pixel and each of the pixels adjacent to the defective pixel with the value of a pixel adjacent to the defective pixel and the pixels adjacent to the defective pixel or the average value of the pixels adjacent to the defective pixel and the pixels adjacent to the defective pixel.

[0020] Embodiments may also relate to an apparatus for correcting defects appearing in an image produced by an image sensor, configured to perform the method defined above.

[0021] Embodiments may also relate to an imaging device configured to perform the method defined above, comprising an image sensor, a circuit for acquiring a temperature of the image sensor, and a circuit for acquiring an integration time applied to the image sensor.

[0022] According to one embodiment, the image sensor is of the CTIA or SF type.

[0023] Embodiments may also relate to a computer program product loadable into the memory of a computer which, when executed by the computer, configures the computer to perform the methods defined above.

[0024] Exemplary embodiments of the present invention will become more clearly apparent from the following description, given for illustrative purposes only and illustrated in the accompanying drawings, in which: [Brief explanation of the drawings]

[0025] [Figure 1] 1 is a schematic representation of a conventional imaging device; [Figure 2] 1 illustrates a schematic representation of an image sensor associated with a noise reduction device for an imaging device according to an embodiment; [Figure 3] 3 illustrates a method for calculating noise reduction factors according to an embodiment; [Figure 4] 1 shows a curve of the variation in intensity of a pixel signal as a function of the temperature of an image sensor. [Figure 5] 1 shows a curve of the variation in intensity of a pixel signal as a function of the temperature of an image sensor. [Figure 6] 1 shows a curve of the variation in intensity of a pixel signal as a function of the temperature of an image sensor. [Figure 7] 2A and 2B schematically represent a block diagram of a gain correction circuit receiving an image from a noise reduction device according to an embodiment. [Figure 8] FIG. 1 illustrates a block diagram of a circuit for generating a table of gain correction factors according to an embodiment; DETAILED DESCRIPTION OF THE INVENTION

[0026] FIG. 1 shows an image sensor IS1 having processing circuits ADC, AMP, and PRC. The image sensor IS1 may be incorporated into a portable device such as a camera, camcorder, mobile phone, or any other device with image capture capabilities. The image sensor IS1 typically includes an array PXA of pixel circuits PC. The array PXA includes pixel circuits PC arranged in multiple rows and multiple columns. The image sensor IS1 also includes control circuits RDRV, RDEC, CDRV, CDEC, and TAC configured to provide different control signals to the pixel circuits PC according to the steps sequenced to capture an image. The readout and processing circuitry may include an amplifier AMP, an analog-to-digital converter ADC, and a processor PRC. The array PXA provides pixel signals to the readout and processing circuits AMP, ADC, and PRC, configured to provide an image IM from the pixel signals.

[0027] Pixel rows are selectively activated by a row driver circuit RDRV in response to a row address decoder RDEC. Pixel columns are then activated by a column driver circuit controlled by a column decoder CDEC. The circuits RDRV and CDRV provide appropriate voltages to drive the pixel circuits PC. The sensor IS also includes a control circuit TAC that drives the address decoders RDEC and CDEC and the driver circuits RDRV and CDRV to select the appropriate pixel row and column for pixel readout at any given time. The readout pixel signals are amplified by an amplifier AMP and converted to digital form by an analog-to-digital converter ADC. The digitally converted pixel signals are processed by an image processor PRC, which derives an image IM from the digitally converted pixel signals. The image processor PRC may include a memory for processing and storing the received image signals.

[0028] 2 illustrates an image sensor IS associated with a noise reduction device according to one embodiment. In the example of FIG. 2, the noise reduction device comprises a noise reduction function performed by a processor IPRC coupled to a memory MEM. The processor IPRC may be integrated into or connected to the image sensor IS. The definition of the noise reduction function is based on a noise model that includes two components: a bias component that varies linearly with the temperature of the image sensor, and a dark current component that varies with both temperature and integration time.

[0029] According to one embodiment, the noise model implemented by the processor IPRC is calculated using the following equation: bm[i,j]=IT×ad[i,j]×Exp(bd[i,j]×TP)+ab[i,j]×TP+bb[i,j] (1) where bm[i,j] is the pixel of the correction image BM to be subtracted from each image generated by the image sensor IS, the pixel location is specified by row and column index i,j, IT is the integration time, TP is the temperature measured by the image sensor IS or a temperature sensor TS coupled to the image sensor, ad[i,j], bd[i,j], ab[i,j] and bb[i,j] are coefficients determined for the pixel to be corrected at position [i,j], and EXP is an exponential function. The coefficients ad[i,j], bd[i,j], ab[i,j] and bb[i,j] are input to the memory MEM of the imager and stored as tables AD, BD, AB, BB. The following equations: px'[i,j]=px(j,j)-bm[i,j] (2) (where px'[i,j] is the corrected value of pixel px[i,j]) to correct each pixel px[i,j] of the image IM generated by the image sensor to generate a corrected image OIM.

[0030] Under these conditions, pixel values ​​provided by the image sensor IS can be individually corrected for all integration times and operating and imager temperatures. Due to the simplicity of the correction, which involves subtracting the corresponding pixel bm[i,j] of the correction image BM from each pixel px[i,j] of the image IM, the images provided by the image sensor IS can be processed at very high frame rates. The processor IPRC can therefore process video streams containing hundreds of frames per second. If necessary, this correction can be performed by hardwired logic circuits (such as those shown in Figure 2). Furthermore, by providing such a circuit for each pixel of a frame, all pixels of each frame can be processed simultaneously in parallel.

[0031] Similarly, the simplicity of equation (1), which contains only addition, multiplication, and exponentiation operations EXP, allows the calculation of the correction image BM to be performed in real time according to changes in temperature TP or integration time IT. Also, if desired, the update function of the correction image BM performed by the processor IPRC may be performed by hardwired logic circuits (such as the circuit shown in Figure 2), and the exponentiation operations EXP may simply be performed by a look-up table. If there is no interaction between the pixels of the image BM, all pixels bm[i,j] of the correction image BM can also be calculated in parallel.

[0032] Figure 3 shows the circuit TGC for generating the coefficient tables AD, BD, AB, and BB. The circuit TGC receives as input dark images DIM[IT0-ITn, TP1-TPm] obtained using an imaging device during a calibration phase by exposing the image sensor IS to different temperatures TP1, TP2, ... TPm and sequentially setting the integration time IT for each temperature to different values ​​IT0, IT1, ... ITn, including the minimum value IT0. The temperatures TP1-TPm can be selected within the operating temperature range of the image sensor IS. Furthermore, the number of integration time values ​​can be set to a number between 5 and 15 within the range of possible integration time values ​​of the imaging device.

[0033] In the first step, bias coefficients ab[i,j], bb[i,j] are calculated from the image DIM[IT0,TP1-TPm] obtained with the minimum integration time IT0, where the integration time IT in equation (1) is a multiplication factor for the dark signal component, and therefore this component becomes negligible when the integration time is very small. According to one example, the minimum integration time IT0 is 10-100 μs, preferably 40-60 μs. As a first approximation, each pixel dx[i,j,IT0,TP] of the image DIM[IT0,TP1-TPm] obtained with the integration time IT0 is calculated using the following equation: dx[i,j,IT0,TP]≒ab[i,j]×TP+bb[i,j] (3) (Where TP=TP1, TP2, ... TPm) Therefore, the bias coefficients ab[i,j] and bb[i,j] from tables AB and BB can be determined for each pixel dx[i,j,IT0,TP] in the temperature range TP1-TPm by a linear regression calculation LR.

[0034] In a second step, the dark coefficients ad[i,j], bd[i,j] of the tables AD, BD are calculated. For this purpose, the corrected image DIM'[IT1-ITn,TP1-TPm] is derived from the image DIM[IT1-ITn,TP1-TPm] by subtracting from each pixel dx[i,j,IT,TP] of each image DIM[IT,TP] obtained for integration times IT1-ITn and temperatures TP1-TPm the value of the corresponding pixel dx[i,j,IT0,TP] located in the image DIM[IT0,TP] obtained for the same temperature TP and minimum integration time IT0. Therefore, taking into account equation (1), each pixel dx'[i,j,IT,TP] of the corrected image DIM[IT1-ITn,TP1-TPm] is calculated using the following equation: dx'[i,j,IT,TP]=IT×ad[i,j,IT]×Exp(bd[i,j,IT]×TP) (4) (wherein dx'[i,j,IT,TP]=dx[i,j,IT,TP]-dx[i,j,IT0,TP]) It may be modeled by:

[0035] The dark coefficients ad[i,j] and bd[i,j] can be determined for each integration time IT1-ITn by an exponential fitting calculation EF. Such calculations are implemented, for example, in the MathWorks® MATLAB software library. The dark coefficients ad[i,j] and bd[i,j] stored in tables AD and BD can be obtained by averaging the coefficients ad[i,j,IT] and bd[i,j,IT] obtained for integration times IT1-ITn, respectively.

[0036] For some image sensors, it may be preferable to add the same compensation component to the noise model for all pixels of the image sensor depending on the integration time. This component can be determined by comparing the pixel at position [i,j] in images DIM obtained at the same temperature TP for integration times IT0-ITn. This comparison can be performed by considering several pixels in each of the images DIM obtained at the same temperature TP for different integration times IT0-ITn, and setting the compensation component to the average value obtained for the pixel considered.

[0037] 4 to 6 show curves C2, C4, and C6 of the variation of the value of pixel px[i,j] as a function of temperature TP in the uncompensated image IM provided by image sensor IS, as well as curves C1, C3, and C5 corresponding to the model bm[i,j] defined by equation (1) and determined for pixel [i,j] by the coefficients ad[i,j], bd[i,j], ab[i,j], and bb[i,j]. Curves C2, C4, and C6 were plotted from pixel values ​​px[i,j] obtained at temperatures TP1≈27°C, TP2≈32°C, TP3≈36.5°C, and TP4≈41°C.

[0038] Curves C1 and C2 in Fig. 4 correspond to an integration time of 50 μs. Fig. 4 shows a difference of less than 0.6% between the value of pixel px[i,j] at the output of the image sensor IS and the corresponding value bm[i,j] determined by the model (equation (1)).

[0039] Curves C3 to C6 in Figures 5 and 6 were obtained at integration times of 6.654 ms and 13.321 ms, respectively. Figure 5 shows a maximum deviation of less than 14% between the measured value of pixel px[i,j] and the corresponding value bm[i,j] determined by the model. Figure 6 shows a maximum deviation of less than 8% between the value of pixel px[i,j] output by image sensor IS and the corresponding value bm[i,j] determined by the model. It can be seen that these maximum deviations are obtained at temperatures above 38°C.

[0040] According to one embodiment, the processor IPRC periodically or continuously receives the temperature TP from the image sensor IS and stores the current temperature measured by the image sensor IS each time the processor IPRC calculates a correction image BM. If the current temperature measurement provided by the image sensor deviates from the stored value, the processor IPRC recalculates the correction image BM taking into account the last temperature measurement provided by the image sensor. Depending on the implementation, a new calculation of the correction image BM is performed when a temperature difference of 0.5°C to 2°C (e.g., 1°C) occurs. Similarly, if the IT integration time is changed, the IPRC processor recalculates the correction image BM according to the new integration time.

[0041] According to one embodiment, the image sensor IS has several gain values ​​that can be selected from the control interface of the imaging device or according to the illumination conditions of the image sensor. In this case, a dark image DIM is generated for each gain value, and a table of coefficients AD, BD, AB, and BB is determined for each gain value. The integration time value IT used to generate the image DIM can be selected based on the gain value to avoid unlikely cases where there is a high risk of saturation of the image sensor IS. Furthermore, the processor IPRC calculates a current correction image BM for each gain value based on the integration time IT and the temperature IS of the image sensor.

[0042] The quality of the images produced by the image sensor can also degrade as a function of the temperature of the image sensor due to pixel-to-pixel gain variations, which can be caused, among other things, by structural differences between pixel circuits.

[0043] 7 shows a gain correction circuit according to one embodiment. The gain correction circuit receives the corrected image OIM at the output of the noise reduction circuit and multiplies each pixel px'[i,j] of the corrected image OIM by a gain correction factor gf[i,j] calculated for pixel [i,j] as a function of the current temperature of the image sensor. In this way, the pixels px''[i,j] of the resulting corrected image GCI are produced with a uniform gain across the entire image GCI.

[0044] The gain correction factors gf[i,j] belong to a gain normalization table GF[i,j] stored in a memory MEM and determined by the interpolation module ITP as a function of the temperature TP of the image sensor, and to a set of gain normalization tables GF[TP1], ... GF[TPm] determined for different temperatures TP1, ... TPm. The interpolation applied by the interpolation module ITP may be, for example, a linear or polynomial interpolation.

[0045] According to one embodiment, the gain normalization table GF is updated when a temperature difference of 0.5°C to 2°C (eg 1°C) occurs with respect to the temperature previously taken into account for the calculation of the correction table.

[0046] The image processing shown in Figure 7 may be performed by a processor IPRC to process video streams containing video streams of hundreds of frames per second. If desired, this processing may be performed by hardwired logic circuits (such as those shown in Figure 7). Furthermore, by providing such circuitry for each pixel of each frame, all pixels of that frame can be processed simultaneously in parallel.

[0047] Similarly, due to the simplicity of the processing, which involves only addition, multiplication, and NOT operations, the calculation of the gain normalization table GF can be performed in real time according to temperature variations. Also, if necessary, the function for updating the gain normalization table GF performed by the processor IPRC may be performed by hardwired logic circuitry (such as the circuit shown in FIG. 7), provided that the interpolation operations can be performed at least in part using one or more lookup tables, if necessary. Also, all pixels gf[i,j] of the gain normalization table GF may be calculated in parallel.

[0048] FIG. 8 illustrates a circuit GGC for generating gain normalization tables GF[TP1], ... GF[TPm] according to one embodiment. This calculation circuit receives as input images UIM[IT0...ITn,TP1...TPm] obtained during a calibration phase using an imaging device placed in front of a light source with an apparently uniform light intensity in all directions. Images UIM[IT0...ITn,TP1...TPm] were obtained by exposing the image sensor IS to different temperatures TP1, ...TPm and sequentially setting the integration time IT to different values ​​IT0, ...ITn for each temperature. The light source used to generate the images UIM[IT0...ITn,TP1...TPm] may be an integrating sphere, and the imaging device is placed within the aperture of the integrating sphere. Instead of changing the integration time IT, it is also possible to vary the intensity of light emitted by the uniform light source by, for example, setting the integration time of the image sensor to an average value. The operating temperatures TP1, ...TPm are set, for example, to 20, 30, 40, and 50°C.

[0049] The circuit GGC calculates a gain table PG[TP] for each temperature TP1-TPm using a linear regression calculation circuit RL, where the gain pg[i,j,TP] for each pixel[i,j] corresponds to the average slope of the curve of the value of pixel px[i,j] as a function of integration time IT. Then, by calculating the average AV, the circuit GGC determines a table of average gains PGM[TP] containing the average gain value for each pixel[i,j] for each temperature TP = TP1, ... TPm. Then, for each pixel[i,j], a respective gain normalization table GF[TP1], ... GF[TPm] is generated by dividing the corresponding value pgm[i,j] in the average gain table PGM[TP] by the corresponding value pg[i,j,TP] in the gain table PG[TP].

[0050] The quality of the image provided by the image sensor IS may be affected by the presence of defective pixel circuits. According to one embodiment, the processor IPRC calculates for each pixel [i,j] an average integration time IT moyis configured to calculate the average deviation E[i,j] at different temperatures TP=TP1, ...TPm between the pixel values ​​dx[i,j,IT,TP] of the image DIM provided by the image sensor IS and the correction value bm[i,j,IT,TP] for this pixel, and compare this average with a threshold. If this average deviation is greater than the threshold for pixel [i,j], pixel [i,j] is considered defective. The detection of defective pixels can be performed during a calibration phase after acquisition of the image DIM. For example, the average deviation E[i,j] can be calculated using the following equation:

number

[0051] According to one embodiment, the processor IPRC corrects the image OIM or GCI by replacing the value px'[i,j] or p''[i,j] of each defective pixel thus detected by the processor IPRC with the value of a neighboring pixel or the average value of the pixels neighboring the defective pixel. The pixels neighboring a defective pixel may also be considered defective and constitute a defective pixel region. In this case, each pixel of the defective pixel region may be replaced by one pixel neighboring the defective pixel region or the average value of these neighboring pixels.

[0052] It will be clear to those skilled in the art that the invention is susceptible to modifications and various applications. In particular, the invention is not limited to an image sensor for calculating the correction image BM. In practice, the correction image BM can be determined by an external computer having tables AB, BB, AD, BD and receiving from the image sensor the integration time IT and the temperature TP of the image sensor IS and, if necessary, the gain applied by the image sensor.

[0053] Additionally, correction of the image provided by the image sensor may also be performed by such an external computer receiving the uncompensated image IM from the image sensor.

[0054] Also, the temperature TP is not necessarily provided by the image sensor IS, but can be measured by an external temperature sensor associated with the image sensor.

[0055] The processed image can still be an image or a video image.

[0056] Just as the gain correction can be performed without the noise reduction correction described with reference to FIG. 2, the pixel gain correction may be performed from a corrected image OIM obtained by another noise reduction process.

[0057] Similarly, the detection of defective pixels as described above can be performed without using any of the correction methods shown in Figures 2 and 7 that are used to obtain images OIM and GCI. Similarly, correction of defective pixels can be performed on uncorrected images, i.e., without using any of the correction methods shown in Figures 2 and 7.

Claims

1. 1. A method for correcting defects appearing in an image produced by an image sensor, the method comprising: receiving an image to be corrected captured by the image sensor; receiving a temperature from the image sensor acquired when the image to be corrected was captured by the image sensor; receiving an integration time (IT) applied by the image sensor when capturing the image to be corrected; subtracting, for each pixel of the image to be corrected, from a pixel value a pixel-specific noise correction factor derived from a noise reduction model including a linear component dependent on the temperature of the image sensor, the linear component being added to a product of an exponential component dependent on the temperature of the image sensor and the integration time, the linear component and the exponential component being determined by coefficients specific to the pixel; A method comprising:

2. The noise reduction model is represented by the following equation: bm[i,j]=IT×ad[i,j]×Exp(bd[i,j]×TP)+ab[i,j]×TP+bb[i,j] where bm[i,j] is a noise correction factor to be subtracted from the corresponding pixel of the image to be corrected, IT is the integration time, TP is the temperature of the image sensor, EXP is an exponential function, and ad[i,j], bd[i,j], ab[i,j] and bb[i,j] are coefficients specific to the pixel. The method of claim 1 , wherein the parameter is defined by:

3. 3. The method of claim 1, further comprising calculating a noise correction factor for each pixel each time the integration time is changed or the temperature of the image sensor deviates from a previous value by more than a temperature deviation threshold.

4. acquiring images with the image sensor in the absence of light at a minimum integration time, each image being taken at a different respective temperature; determining the coefficients of the linear components for each pixel of the image to be corrected by a linear regression calculation applied to corresponding pixels in an image taken in the absence of light at each temperature; 4. The method of claim 1, comprising:

5. acquiring images with the image sensor in the absence of light at different integration times, exposing the image sensor to different temperatures; generating corrected images by subtracting an image taken by the image sensor in the absence of light at the minimum integration time and at the same temperature from each of images taken by the image sensor in the absence of light at different integration times and at different temperatures; determining the coefficients of the exponential component by an exponential fitting calculation applied to each pixel of the corrected images obtained at different temperatures and corresponding to the same integration time; The method of claim 4, comprising:

6. The method of claim 5 , wherein the coefficients of the exponential component are determined by averaging coefficients obtained by exponential fitting calculations for different integration times.

7. 7. The method of claim 1, wherein the noise reduction model comprises components that are identical for all pixels of the image sensor depending on the integration time.

8. 8. A method according to any one of claims 1 to 7, comprising a step of acquiring a video stream image, wherein a noise correction factor corresponding to each pixel of the image of the video stream is subtracted from the corresponding pixel of each image of the video stream.

9. receiving a command to select a gain value for the image sensor; selecting a pixel-specific noise correction factor for each pixel of the image to be corrected as a function of the selected gain value, the noise correction factor being used to correct a value of each pixel of the image to be corrected, the noise correction factor being determined from a set of pixel-specific coefficients generated as a function of the selected gain value; 9. The method of claim 1, comprising:

10. 10. The method of claim 1, comprising, for each pixel of the corrected image after noise reduction, multiplying the value of the pixel by a gain correction factor specific to that pixel selected from a gain normalization table to obtain an image with normalized gain.

11. 11. The method of claim 10, further comprising updating the gain normalization table each time the integration time is changed or each time a temperature of the image sensor deviates from a previous value by more than a temperature deviation threshold, wherein updating the gain normalization table is performed by an interpolation calculation applied to a set of gain normalization tables determined for different temperatures.

12. acquiring a series of images with the image sensor in the presence of a uniform light source at different integration times or different light source intensities, each series of images being taken at a different respective temperature; determining a gain for each pixel in one of the images of each series of images by a linear regression calculation applied to corresponding pixels in the images of the series of images; determining, for each pixel of one image of each series of images, a gain correction factor by dividing the average of gains obtained for all pixels of the series of images by the gain determined for the pixel; The method of claim 11 , comprising:

13. acquiring a series of images with the image sensor in the absence of light over an average integration time, each image being obtained with the image sensor exposed to a respective temperature; calculating, for each pixel of an image in the series of images acquired, an average deviation at different image sensor temperatures, each deviation being calculated for one image sensor temperature between the value of the pixel in the image in the series of images corresponding to the image sensor temperature and a noise correction factor determined for the pixel at the average integration time and the image sensor temperature; comparing the mean deviation to a threshold and deeming the pixel defective if the mean deviation is greater than the threshold for the pixel; 13. The method of any one of claims 1 to 12, comprising:

14. correcting each image captured by the image sensor by replacing the value of a defective pixel with the value of an adjacent pixel or the average value of adjacent pixels; or correcting each image captured by the image sensor by replacing the value of each of a defective pixel and a plurality of pixels adjacent to the defective pixel with the value of a pixel adjacent to the defective pixel and a plurality of pixels adjacent to the defective pixel or with an average value of a plurality of pixels adjacent to the defective pixel and a plurality of pixels adjacent to the defective pixel.

14. The method of claim 13, comprising:

15. Apparatus for correcting defects appearing in images produced by an image sensor, configured to carry out a method according to any one of claims 1 to 14.

16. 15. An imaging device configured to perform the method of any one of claims 1 to 14, comprising an image sensor, a circuit for obtaining a temperature of the image sensor, and a circuit for obtaining an integration time applied to the image sensor.

17. 17. Apparatus according to claim 15 or 16, wherein the image sensor is of the CTIA or SF type.

18. A computer program loadable into the memory of a computer, which, when executed by said computer, configures said computer to carry out the method of any one of claims 1 to 14.

Images