Device comprising a camera sensor
By generating feature vectors to identify strip frequencies and adjusting the integration time of the camera sensor, the problem of uneven image brightness under periodic light sources was solved, and image acquisition with uniform brightness was achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- STMICROELECTRONICS FRANCE
- Filing Date
- 2022-07-08
- Publication Date
- 2026-05-22
AI Technical Summary
When using periodic light sources, existing camera sensors are prone to producing bands of varying brightness in images, causing the images to not correspond to reality. Existing solutions are either costly or aesthetically unappealing.
By generating feature vectors to identify the frequency of bands in an image, the integration time of the camera sensor is adjusted to match the period of the light pulse from the light source, thus eliminating the bands.
It achieves the precise elimination of different brightness bands in an image without increasing costs, resulting in an image with uniform brightness, unaffected by the reflectivity of scene objects and light levels.
Smart Images
Figure CN115604591B_ABST
Abstract
Description
[0001] Cross-references to related applications
[0002] This application claims the benefit of French patent application No. 2107444, filed on July 9, 2021, which is incorporated herein by reference. Technical Field
[0003] This invention relates to a camera device, and more particularly to a camera device having a rolling shutter camera sensor. Background Technology
[0004] A rolling shutter camera sensor includes a matrix of photodetectors arranged in columns and rows. The camera sensor is configured to acquire an image by performing line-by-line acquisition of signals emitted by the photodetectors. This type of acquisition is called a rolling shutter.
[0005] Specifically, a rolling shutter scan of the photodetector rows of the camera sensor is performed at a given frequency. Therefore, the rows of the image are not acquired precisely at the same time. For each row, the rolling shutter scan consists of an exposure phase performed within a given integration time, followed by a readout phase. During the exposure phase, the photodetectors of the rows detect the light they receive. Then, during the readout phase, the amount of light detected by the photodetectors is read out.
[0006] Furthermore, there are light sources controlled by a pulse-width modulated control signal (also known as PWM). For example, a light source with a light-emitting diode (LED) is configured to control the emission of the LED by a pulse-width modulated signal. In particular, such a light source is controlled by pulses of a control signal emitted periodically according to a given frequency.
[0007] The intensity of light emitted by this source increases during the pulse of the control signal and then decreases after the pulse until the next pulse. Therefore, the intensity of light emitted by this light source varies with time. Specifically, this light source emits light pulses periodically. The period of the light pulses from this source corresponds, for example, to the duration between the two rising edges of two consecutive pulses.
[0008] A camera sensor can acquire images illuminated by this light source. However, the period of the light pulses from the light source can differ from the integration time of each line of the camera sensor.
[0009] Therefore, the light intensity sensed during the acquisition of one row of photodetectors on the imaging sensor may be different from the light intensity sensed during the acquisition of multiple rows of photodetectors before or after the current acquisition, which are time-shifted from the current acquisition.
[0010] This has the effect of producing stripes with varying brightness in the acquired image. Therefore, the acquired image does not correspond to perceived reality. Summary of the Invention
[0011] The embodiments provide an imaging device and an imaging sensor. Various embodiments provide stripes with substantially the same brightness in the acquired image (e.g., from multiple rows of photodetectors). Various other embodiments provide stripes with reduced or eliminated brightness differences in the acquired image.
[0012] For example, camera sensors have been proposed with integration time that is a multiple of the period of the light pulse from the light source. Therefore, the exposure duration of the photodetector allows for receiving the same average light intensity. However, the frequency of the light pulse from the light source is not normalized. Consequently, such camera sensors do not allow for the elimination of stripes with varying brightness for light sources having light pulse frequencies different from those with adapted integration times.
[0013] Furthermore, a method has been proposed to perform subtraction between two consecutively acquired images by eliminating bands with different brightness levels detected before different brightness levels are detected. However, this solution becomes unapplicable once an object or person moves within the scene between the two images used to perform this subtraction.
[0014] An approach to adding an ambient light sensor has also been proposed, which involves detecting the frequency of the light pulses from the light source before adjusting the integration time of the camera sensor based on the measured frequency of the light pulses. However, this solution is expensive and requires a location within the device that includes the camera sensor. This additional location can be considered aesthetically unpleasant.
[0015] Therefore, there is a need for a simple and cost-effective solution for image acquisition that allows for the acquisition of images of scenes illuminated by artificial light sources that do not have stripes of varying brightness independent of the frequency of the light pulses from the light source.
[0016] According to one embodiment, an apparatus is provided comprising:
[0017] - A scanning camera sensor is configured to acquire an image of a scene illuminated by a source with regular light pulses, the light pulses being emitted periodically according to a given integration time, such that when the integration time of the sensor differs from the period of the light pulses, the acquired image can have regular continuous bands containing different brightness levels.
[0018] - The processing unit is configured as follows:
[0019] The system generates regular, continuous feature vectors representing bands of varying brightness present in an image acquired by the camera sensor. These feature vectors are independent of the reflectivity of objects in the scene being captured and the level of light in the scene.
[0020] The frequency of the bands in the image is determined based on the generated feature vector.
[0021] The period of the source pulse is determined based on the frequency of the bands in the determined image.
[0022] - A control unit is configured to adjust the integration time of the camera sensor such that the integration time is a multiple of the period of the determined light pulse of the source.
[0023] Because the feature vector is independent of the reflectivity of objects in the scene being photographed and the light level in the scene, it is easy to identify bands with different brightness levels within the feature vector. This allows for the precise determination of the frequencies of these bands with different brightness levels, and thus, the precise determination of the frequencies of the light pulses emitted by the light source.
[0024] By adjusting the integration time of the camera sensor according to the determined frequencies of stripes with different brightness, a scan of the photodetector rows of the camera sensor is performed, ensuring that the photodetector rows are exposed to the same average light intensity. Therefore, the resulting image does not contain stripes with different brightness.
[0025] Such a device therefore allows the elimination of bands with different brightness in the image obtained after adjustment.
[0026] This device does not require additional sensors, such as ambient light sensors, to determine the frequency of the light pulses from the light source.
[0027] The acquired image is composed of pixels with individual color components.
[0028] In an advantageous embodiment, in order to generate feature vectors associated with the color components of the image, the processing unit is configured to:
[0029] - For each row of the acquired image:
[0030] The values of the pixels with color components in that row are summed to obtain the values representing that color component for that row.
[0031] The value representing the color component is divided by the value representing the same color component in the adjacent row of the image.
[0032] The values of pixels with the same color component in each row are summed to obtain a representation that allows for improved robustness to noise.
[0033] This division allows the feature vector to be independent of the reflectivity of objects in the scene being photographed and the light level in the scene.
[0034] However, in an advantageous embodiment, in order to generate feature vectors associated with the color components of the image, the processing unit is configured to:
[0035] - For each row of the acquired image:
[0036] The values of the pixels with color components in that row are summed to obtain the values representing that color component for that row.
[0037] Calculate the logarithm of the value representing the color component of that row.
[0038] This function distinguishes between the logarithmic value representing the value of the color component in that row and the logarithmic value representing the value of the same color component in adjacent rows of the image.
[0039] Entering the logarithmic field allows for simplified computation by the processing unit to generate a feature vector relating to the division performed in the foregoing embodiments. Furthermore, the use of the logarithmic field allows for a reduction in the memory size required to record the results through the natural compression effect of logarithms.
[0040] Preferably, in order to generate a feature vector associated with the brightness of the image, the processing unit is configured to:
[0041] - For each row of the acquired image:
[0042] The values of pixels with the same color component in that row are summed to obtain a representation value for each color component in that row.
[0043] The value representing the brightness of a row is calculated based on the values of the color components representing the same row and optionally adjacent rows.
[0044] Calculate the logarithm of the value representing the brightness of that row.
[0045] о Calculate the difference between the logarithm of the value representing the brightness of the current row and the logarithm of the value representing the brightness of the adjacent row of the image.
[0046] The feature vectors associated with brightness have the advantage of high signal-to-noise ratio.
[0047] Advantageously, the scanning camera sensor is configured to acquire a series of images, each of which can have regular, continuous stripes. Therefore, the acquired image can consist of pixels with individual color components.
[0048] In an advantageous embodiment, in order to generate feature vectors associated with the color components of the image, the processing unit is configured to:
[0049] - For each row of the acquired image:
[0050] The values of the pixels that have the color component in that row are summed to obtain the values representing that color component for that row.
[0051] The value representing the color component is divided by the value of the same color component representing the same row in a subsequent or previous image of a consecutive image.
[0052] However, in an advantageous embodiment, in order to generate feature vectors associated with the color components of the image, the processing unit is configured to:
[0053] - For each row of the acquired image:
[0054] The values of the pixels that have the color component in that row are summed to obtain the values representing that color component for that row.
[0055] Calculate the logarithm of the value representing the color component of that row.
[0056] The difference between the logarithm of the value of the color component representing the row and the logarithm of the value of the same color component representing the same row in a subsequent or previous image in a series of images is calculated.
[0057] Preferably, in order to generate a feature vector associated with the brightness of the image, the processing unit is configured to:
[0058] - For each row of the acquired image:
[0059] The values of pixels with the same color component in that row are summed to obtain a representation value for each color component in that row.
[0060] The value representing the brightness of a row is calculated based on the values of the individual color components representing the same row and adjacent rows.
[0061] Calculate the logarithm of the value representing the brightness of that row.
[0062] The function performs the difference between the logarithm of the value representing the brightness of the row and the logarithm of the value representing the brightness of the same row in a subsequent or consecutive image.
[0063] According to another embodiment, a method for adjusting the integration time of a scanning camera sensor is proposed, comprising:
[0064] - An image of a scene illuminated by a source with regular light pulses is acquired based on a given integration time. The light pulses are periodic according to a given period, such that when the integration time of the sensor differs from the period of the light pulses, the acquired image can have regular, continuous stripes containing different brightness levels.
[0065] - Generate feature vectors representing regular, continuous stripes of varying brightness present in an image acquired by the camera sensor, wherein the feature vectors are independent of the reflectivity of objects in the captured scene and the level of light in the scene.
[0066] - Determine the frequency of bands in the image based on the generated feature vectors.
[0067] - Determine the period of the source pulses based on the frequencies of the stripes in the determined image.
[0068] - Adjust the integration time of the camera sensor so that the integration time is a multiple of the period of the light pulse from the source. Attached Figure Description
[0069] Other advantages and features of the invention will become apparent upon examination of the detailed description of the embodiments and implementations (in no way limiting) and the accompanying drawings, in which:
[0070] Figure 1 An apparatus according to an embodiment is shown;
[0071] Figure 2 The exposure phases of each row of photodetectors are shown by time offset;
[0072] Figure 3 An example of an image obtained by a camera sensor is shown;
[0073] Figure 4 A method for adjusting the integration time of a camera sensor to reduce or eliminate differences in brightness in different bands of the acquired image is shown.
[0074] Figure 5 A method for generating feature vectors obtained according to the first embodiment is shown;
[0075] Figure 6 A method for generating feature vectors of acquired images according to a second embodiment is shown;
[0076] Figure 7 A method for generating feature vectors of acquired images according to a third embodiment is shown;
[0077] Figure 8 A method for generating feature vectors of acquired images according to a fourth embodiment is shown; and
[0078] Figure 9 A schematic diagram of the feature vector VCS according to an embodiment is shown. Detailed Implementation
[0079] Figure 1 A device APP according to an embodiment is shown. The device APP includes a camera sensor CPH, a control unit UCO, and a processing unit UT.
[0080] The camera sensor CPH comprises a matrix of photodetectors MPHDs. The photodetectors PHDs of the MPHD matrix are arranged in rows and columns. The photodetectors PHDs are associated with different color components, particularly with the red (R), green (Gr / Gb), and blue (B) components. For example, the photodetectors PHDs can be organized according to a Bayer matrix known to those skilled in the art.
[0081] The camera sensor CPH is configured to acquire an image of the scene by acquiring a signal generated by the photodetector PHD based on the light sensed by the photodetector PHD.
[0082] Specifically, the acquisition of signals generated by the photodetector PHD is performed line by line. In other words, the camera sensor CPH allows images to be acquired by scanning the lines of the photodetector PHD of the sensor CPH.
[0083] The scanning causes a time-related acquisition offset of the signals from the photodetector PHDs of each row of photodetectors.
[0084] In particular, such as Figure 2 As shown, the acquisition of a row of photodetectors (PHDs) includes an exposure phase (PEXP), where the photodetectors in that row detect the light they receive during a given integration time. The acquisition then includes a readout phase (PLEC), where the amount of light detected by the photodetectors (PHDs) is read out.
[0085] The photodetector line PHD is scanned according to the scanning frequency defined by the control unit UCO of the device APP.
[0086] Figure 2 The exposure phases of the PEXP for each row of the photodetector PHD are shown with temporal offset.
[0087] Furthermore, the scene being filmed can be illuminated by a source with regular light pulses. For example, this source could be a source with light-emitting diodes (LEDs).
[0088] This type of light source emits light pulses at a given frequency.
[0089] like Figure 2 As shown, the intensity ISRC of the light emitted by this light source is maximum at the moment of pulse IMP.
[0090] The integration time of the camera sensor cannot be a multiple of the period of the light pulse from the light source illuminating the scene.
[0091] Therefore, the exposure phase PEXP of each row of photodetectors can be offset relative to the light pulse IMP of the light source. Consequently, the intensity ISRC of the light detected by each row can vary. For example... Figure 3 As shown, the image IMG obtained by the camera sensor therefore has bands BDL containing different brightness levels.
[0092] The device app is configured to implement a method for adjusting the integration time of the camera sensor in order to eliminate bands with varying brightness in the acquired image. This adjustment method... Figure 4 As shown in the image.
[0093] The adjustment method includes step 40, in which the camera sensor CPH uses regular light pulses to acquire at least one image (IMG) of the scene illuminated by the light source. Specifically, the camera device can acquire continuous images.
[0094] Then, in step 41, at least one image is sent to the processing unit.
[0095] Then, in step 42, the processing unit UT generates at least one feature vector VCS based on at least one acquired image IMG. This at least one feature vector is generated in such a way that it can distinguish bands with different brightness levels in the image IMG, regardless of the reflectivity of objects in the captured scene or the light level in the scene. Preferably, the processing unit UT is configured to generate the feature vector VCS for each color component of the image.
[0096] Then, in step 43, the processing unit UT determines the frequencies of stripes with different brightness based on at least one feature vector VCS. The frequencies of the stripes with different brightness depend on the frequency of the light pulse and the frequency of the photodetector rows of the scanning sensor according to the following formula:
[0097]
[0098] Where f bandes f is the frequency of the bands with different brightness in each row of the image. balayage It is the scanning frequency of the sensor's photodetector over time, and f impulsions It is the frequency of the light pulse over time.
[0099] The processing unit UT then sends the calculated frequency of the light pulses from the light source to the control unit UCO.
[0100] Finally, in step 44, the control unit UCO adjusts the integration time of the camera sensor so that the integration time is a multiple of the light pulse period of the light source.
[0101] Therefore, bands with different brightness are eliminated in the next acquired NIMG image because the average light intensity remains constant during the exposure phase of each row of the photodetector of the camera sensor.
[0102] More specifically, the processing unit can implement various methods to obtain feature vectors.
[0103] Figure 5 A first embodiment of a method for generating feature vectors (VCS) of acquired images (IMGs) in isolation or sequentially is shown.
[0104] In this first embodiment, in step 50, the processing unit UT adds the values of pixels with the same color component of the same row of the image IMG to each row of the image, so as to obtain the row value representing the color component for each row.
[0105] Then, in step 51, the processing unit UT performs a division for each row of the image and for each color component, between the value of the color component representing that row and the value of the same color component representing the adjacent row of the same image. Alternatively, for each row of the image and for each color component, the processing unit performs a division between the value of the color component representing that row and the value of the same color component representing the same row of a previous image in a sequence of acquired images, or a division between the values of the same color component in a subsequent image of the acquired image sequence.
[0106] The division performed allows us to obtain the feature vector (VCS) for each color component of the image. Each feature vector associated with a color component includes the result of the division performed on each row of the image for that color component.
[0107] The feature vector VCS obtained through the above division has the advantage of being independent of the reflectivity of the scene objects and the light level in the scene. Therefore, the feature vector VCS allows for the primary manifestation of brightness variations between two bands with different brightness levels appearing in the generated image. Thus, detecting bands with different brightness levels is simpler based on such a feature vector.
[0108] However, this division involves significant costs in terms of computing resources.
[0109] To simplify the calculations that must be performed by the processing unit UT, it is advantageous to enter the logarithmic field to avoid performing division between the represented values.
[0110] in this regard, Figure 6A second embodiment of a method for generating a feature vector VCS for an image IMG, which is an image acquired in isolation or continuously, is shown.
[0111] In this second embodiment, in step 60, the processing unit UT adds the values of pixels with the same color component to each row of the image so as to obtain the value of that row representing that color component for each row.
[0112] Then, for each row and for each color component, the processing unit UT calculates the logarithm representing the value of that color component in that row in step 61. Specifically, the processing unit calculates the binary logarithm of each value representing the color component in that row.
[0113] Then, in step 62, the processing unit UT calculates, for each row of the image and for each color component, the difference between the logarithm of the value of the color component in that row representing the image and the logarithm of the value of the same color component in the adjacent row representing the image. Alternatively, the processing unit UT may calculate, for each row of the image and for each color component, the difference between the logarithm of the value of the color component in that row representing the image and the logarithm of the value of the same color component in the same row of the preceding or following image in the acquired series of images.
[0114] The subtraction performed allows the acquisition of a feature vector (VCS) for each color component of the image. Each feature vector associated with a color component includes the result of the subtraction performed on each row of the image for that color component.
[0115] Therefore, in this embodiment, the processing unit simply performs subtraction in the logarithmic field instead of division. This simplifies the computation of the processing unit UT.
[0116] Figure 7 A third embodiment of a method for generating a feature vector VCS for an image IMG, which is an image acquired in isolation or continuously, is shown.
[0117] In this third embodiment, in step 70, the processing unit UT counts the number of pixels to be considered for each row and for each color component to generate a feature vector for that color component. For example, the processing unit may consider all pixels in a row that have the same color component.
[0118] Alternatively, the processing unit can consider only a subset of pixels in a row that have the same color components. This allows for the exclusion of certain pixels with outliers, generated by the saturation of the photodetector that produced the outlier, or by marker outlines in the captured scene.
[0119] Then, in step 71, the processing unit UT calculates the logarithm of the number of pixels to be considered. Specifically, the processing unit UT calculates the binary logarithm of the number of pixels to be considered.
[0120] In step 72, the processing unit UT also adds the values of pixels with the same color component in the same row of the image to be considered for each row and for each color component, so as to obtain the row value representing that color component for each row.
[0121] Then, in step 73, for each row and for each color component, the processing unit UT calculates the logarithm representing the value of that color component in that row. Specifically, the processing unit UT calculates the binary logarithm representing each value of the color component in that row.
[0122] Then, in step 74, the processing unit UT calculates a normalized logarithmic value for each row and for each color component by subtracting the logarithmic value representing the value of the color component from the logarithmic value of the number of pixels to be considered. Therefore, the normalized logarithmic value is analogous to the average value of pixels with the same color component in the same row of the image to be considered.
[0123] Then, in step 75, the processing unit calculates the difference between, for each row of the image and for each color component, a normalized logarithmic value associated with that color component in that row of the image and a normalized logarithmic value associated with the same color component in an adjacent row of the image. Alternatively, the processing unit may calculate the difference between, for each row of the image and for each color component, a normalized logarithmic value associated with that color component in that row of the image and a normalized logarithmic value associated with the same color component in the same row of a previous or subsequent image in the acquired series of images.
[0124] These differences, performed by the processing unit, allow the acquisition of a feature vector (VCS) for each color component of the image. Each VCS associated with a color component comprises the result of the difference operation performed on each row of the image for that color component.
[0125] Figure 8 A fourth embodiment of a method for generating a feature vector VCS for an image IMG, which is an image acquired in isolation or continuously, is shown.
[0126] In this fourth embodiment, in step 80, the processing unit UT counts the number of pixels to be considered for each row and for each color component to generate a feature vector for that color component.
[0127] In step 81, the processing unit UT also adds the values of pixels with the same color component in the same row of the image to be considered for each row and for each color component, so as to obtain the row value representing that color component for each row.
[0128] Then, in step 82, for each row, the processing unit UT replaces the value representing the green component of that row with the value representing the brightness Y of that row. In the case of a Bayer matrix, the value representing the brightness of the row is calculated by a weighted average of the values representing the color components of that row and optionally adjacent rows.
[0129] Then, in step 83, for each row, and for the red and blue components, and for the luminance, the processing unit UT calculates the logarithm representing the value of that color component or the luminance in that row. Specifically, the processing unit UT calculates the binary logarithm representing each value of the red and blue components and the luminance in that row.
[0130] In step 84, the processing unit UT also calculates the logarithm of the number of pixels to be considered. Specifically, the processing unit calculates the binary logarithm of the number of pixels to be considered.
[0131] Then, in step 85, the processing unit UT calculates normalized logarithms for each row, for the red and blue components, and for the luminance by performing the difference between the logarithm of the value representing the color component or luminance and the logarithm of the number of pixels to be considered.
[0132] Then, in step 86, the processing unit UT calculates the difference between the normalized logarithmic value associated with the color component of that row of the image and the normalized logarithmic value associated with the same color component of the adjacent row of the image for each row of the image, as well as for the red and blue components. The processing unit UT also calculates the difference between the normalized logarithmic value associated with the brightness of that row of the image and the normalized logarithmic value associated with the brightness of the adjacent row of the image for each row of the image.
[0133] Alternatively, the processing unit UT can calculate, for each row of the image and for the red and blue components, the difference between the normalized logarithmic value associated with that color component in that row of the image and the normalized logarithmic value associated with the same color component in the same row of a previous or subsequent image in a series of images. Therefore, the processing unit UT also calculates, for each row of the image, the difference between the normalized logarithmic value associated with the brightness of that row of the image and the normalized logarithmic value associated with the brightness of the same row of a previous or subsequent image in a series of images.
[0134] These subtractions performed by the processing unit UT allow the acquisition of the feature vectors VCS of the red and blue components and the image brightness.
[0135] The feature vectors associated with brightness have the advantage of high signal-to-noise ratio.
[0136] Figure 9 A graphical representation of the feature vector VCS obtained through the above embodiments is shown.
[0137] The feature vector VCS has the values V_VCS for each row of LIMG in the image acquired by the camera sensor. For example... Figure 9 As shown, the value V_VCS varies only for rows in the image where there are bands with different brightness levels at their edges. These variations occur in... Figure 9 The value in the middle is in the form of the peak amplitude PC.
[0138] The frequencies of stripes with different brightness can therefore be extracted from the feature vector VCS in order to perform an adjustment of the integration time of the camera sensor to eliminate stripes with different brightness in a later-acquired image of the scene, as described above.
[0139] Although the invention has been described with reference to illustrative embodiments, this description is not intended to be limiting. Various modifications and combinations of the illustrative embodiments, as well as other embodiments of the invention, will be apparent to those skilled in the art based on the description. Therefore, the appended claims are intended to cover any such modifications or embodiments.
Claims
1. An apparatus for adjusting the integration time of a scanning camera sensor, comprising: The scanning camera sensor is configured to acquire an image of a scene illuminated by a source with periodically emitted light pulses, based on the integration time of the sensor, such that when the integration time of the sensor is different from the period of the light pulses, the image has regular continuous stripes containing different brightness. The processor is configured as follows: For the color components of the image, a feature vector is generated representing the regular continuous stripes with different brightness present in the image acquired by the camera sensor, wherein the feature vector is independent of the reflectivity of objects in the scene and the light level in the scene; Based on the generated feature vector, the frequency of the band in the image is determined; as well as Based on the determined frequency of the strip in the image, the period of the pulse of the source is determined; as well as The controller is configured to adjust the integration time of the camera sensor such that the integration time is a multiple of the determined period of the light pulse of the source; For each row of the image, the processor is configured as follows: To obtain the values of the color components representing the row, the values of the pixels in the row having the color components are summed, and The value of the color component representing the row is then divided by the value of the same color component representing the adjacent row of the image to generate the feature vector for each of the multiple color components of the image.
2. The apparatus of claim 1, wherein the scanning camera sensor is configured to acquire continuous images, each of the continuous images comprising continuous stripes.
3. An apparatus for adjusting the integration time of a scanning camera sensor, comprising: The scanning camera sensor is configured to acquire an image of a scene illuminated by a source with periodically emitted light pulses, based on the integration time of the sensor, such that when the integration time of the sensor is different from the period of the light pulses, the image has regular continuous stripes containing different brightness. The processor is configured as follows: For the color components of the image, a feature vector is generated representing the regular continuous stripes with different brightness present in the image acquired by the camera sensor, wherein the feature vector is independent of the reflectivity of objects in the scene and the light level in the scene; Based on the generated feature vector, the frequency of the band in the image is determined; as well as Based on the determined frequency of the strip in the image, the period of the pulse of the source is determined; as well as The controller is configured to adjust the integration time of the camera sensor such that the integration time is a multiple of the determined period of the light pulse of the source; For the rows of the image, the processor is further configured to: To obtain the values of the color components representing the row, the values of the pixels in the row having the color components are summed. Calculate the logarithm of the value representing the color component of the row, and The difference between the logarithm of the value of the color component representing the row and the logarithm of the value of the same color component representing the adjacent row of the image is calculated to generate the feature vector for each of the plurality of color components of the image.
4. An apparatus for adjusting the integration time of a scanning camera sensor, comprising: The scanning camera sensor is configured to acquire an image of a scene illuminated by a source with periodically emitted light pulses, based on the integration time of the sensor, such that when the integration time of the sensor is different from the period of the light pulses, the image has regular continuous stripes containing different brightness. The processor is configured as follows: For the color components of the image, a feature vector is generated representing the regular continuous stripes with different brightness present in the image acquired by the camera sensor, wherein the feature vector is independent of the reflectivity of objects in the scene and the light level in the scene; Based on the generated feature vector, the frequency of the band in the image is determined; as well as Based on the determined frequency of the strip in the image, the period of the pulse of the source is determined; as well as The controller is configured to adjust the integration time of the camera sensor such that the integration time is a multiple of the determined period of the light pulse of the source; The feature vector is generated as a feature vector associated with the brightness of the image, and For the rows of the image, the processor is further configured to: To obtain the representation value of each color component of the row, the values of pixels in the row with the same color component are summed. Based on the values of the individual color components representing the same row, a value representing the brightness of the row is calculated. Calculate the logarithm of the value representing the brightness of the row, and The difference between the logarithm of the value representing the brightness of the row and the logarithm of the value representing the brightness of the adjacent row of the image is calculated to generate the feature vector for each of the multiple color components of the image.
5. The apparatus of claim 4, wherein the value representing the brightness of the row is calculated based on the values representing the respective color components of the same row and adjacent rows.
6. An apparatus for adjusting the integration time of a scanning camera sensor, comprising: The scanning camera sensor is configured to acquire an image of a scene illuminated by a source with periodically emitted light pulses, based on the integration time of the sensor, such that when the integration time of the sensor is different from the period of the light pulses, the image has regular continuous stripes containing different brightness. The processor is configured as follows: For the color components of the image, a feature vector is generated representing the regular continuous stripes with different brightness present in the image acquired by the camera sensor, wherein the feature vector is independent of the reflectivity of objects in the scene and the light level in the scene; Based on the generated feature vector, the frequency of the band in the image is determined; as well as Based on the determined frequency of the strip in the image, the period of the pulse of the source is determined; as well as The controller is configured to adjust the integration time of the camera sensor such that the integration time is a multiple of the determined period of the light pulse of the source; For the rows of the image, the processor is further configured to: To obtain the values of the color components representing the row, the values of the pixels in the row having the color components are summed, and The value representing the color component is divided by the value representing the same color component in the same row of a subsequent or previous image in a series of images to generate the feature vector for each of the multiple color components of the image.
7. An apparatus for adjusting the integration time of a scanning camera sensor, comprising: The scanning camera sensor is configured to acquire an image of a scene illuminated by a source with periodically emitted light pulses, based on the integration time of the sensor, such that when the integration time of the sensor is different from the period of the light pulses, the image has regular continuous stripes containing different brightness. The processor is configured as follows: For the color components of the image, a feature vector is generated representing the regular continuous stripes with different brightness present in the image acquired by the camera sensor, wherein the feature vector is independent of the reflectivity of objects in the scene and the light level in the scene; Based on the generated feature vector, the frequency of the band in the image is determined; as well as Based on the determined frequency of the strip in the image, the period of the pulse of the source is determined; as well as The controller is configured to adjust the integration time of the camera sensor such that the integration time is a multiple of the determined period of the light pulse of the source; For the rows of the image, the processor is further configured to: To obtain the values of the color components representing the row, the values of the pixels in the row having the color components are summed. Calculate the logarithm of the value representing the color component of the row, and The difference between the logarithm of the value of the color component representing the row and the logarithm of the value of the same color component representing the same row in a subsequent or previous image in a series of images is calculated to generate the feature vector for each of the multiple color components of the image.
8. An apparatus for adjusting the integration time of a scanning camera sensor, comprising: The scanning camera sensor is configured to acquire an image of a scene illuminated by a source with periodically emitted light pulses, based on the integration time of the sensor, such that when the integration time of the sensor is different from the period of the light pulses, the image has regular continuous stripes containing different brightness. The processor is configured as follows: For the color components of the image, a feature vector is generated representing the regular continuous stripes with different brightness present in the image acquired by the camera sensor, wherein the feature vector is independent of the reflectivity of objects in the scene and the light level in the scene; Based on the generated feature vector, the frequency of the band in the image is determined; as well as Based on the determined frequency of the strip in the image, the period of the pulse of the source is determined; as well as The controller is configured to adjust the integration time of the camera sensor such that the integration time is a multiple of the determined period of the light pulse of the source; The feature vector is generated as a feature vector associated with the brightness of the image, and For the rows of the image, the processor is further configured to: To obtain the representation value of each color component of the row, the values of pixels in the row with the same color component are summed. Based on the values of each color component representing the same row and adjacent rows, a value representing the brightness of the row is calculated. Calculate the logarithm of the value representing the brightness of the row, and The difference between the logarithm of the value representing the brightness of the row and the logarithm of the value representing the brightness of the same row in a subsequent or previous image in a series of images is calculated to generate the feature vector for each of the multiple color components of the image.
9. A method for adjusting the integration time of a scanning camera sensor, comprising: The scanning camera sensor acquires an image of a scene illuminated by a source emitting periodic light pulses based on the integration time, such that when the integration time is different from the period of the light pulses, the image has regular continuous stripes with different brightness. The processor generates feature vectors representing regular continuous stripes with different brightness for the color components of the image, the feature vectors being independent of the reflectivity of objects in the scene and the light level in the scene; The processor determines the frequency of the bands in the image based on the generated feature vector; The processor determines the period of the pulse from the source based on the frequency determined by the stripe in the image; as well as The integration time of the camera sensor is adjusted by the controller so that the integration time is a multiple of the period of the light pulse of the source; The feature vector is generated as a feature vector associated with the color components of the image, and For rows of the image, the method further includes: The processor sums the values of the pixels having the color components in the row in a manner that represents the values of the color components of the row. The processor then divides the value representing the color component by the value representing the same color component in adjacent rows of the image to generate the feature vector for each of the multiple color components of the image.
10. The method of claim 9, wherein the scanning camera sensor acquires continuous images, each of the continuous images comprising continuous stripes.
11. A method for adjusting the integration time of a scanning camera sensor, comprising: The scanning camera sensor acquires an image of a scene illuminated by a source emitting periodic light pulses based on the integration time, such that when the integration time is different from the period of the light pulses, the image has regular continuous stripes with different brightness. The processor generates feature vectors representing regular continuous stripes with different brightness for the color components of the image, the feature vectors being independent of the reflectivity of objects in the scene and the light level in the scene; The processor determines the frequency of the bands in the image based on the generated feature vector; The processor determines the period of the pulse from the source based on the frequency determined by the stripe in the image; as well as The integration time of the camera sensor is adjusted by the controller so that the integration time is a multiple of the period of the light pulse of the source; The feature vector is generated as a feature vector associated with the color components of the image, and For rows of the image, the method further includes: The processor sums the values of the pixels in the row that have the color components in a manner that represents the values of the color components in the row. The processor calculates the logarithmic value representing the color component of the row. The processor calculates the difference between the logarithm of the value of the color component representing the row and the logarithm of the value of the same color component representing the adjacent row of the image, to generate the feature vector for each of the plurality of color components of the image.
12. A method for adjusting the integration time of a scanning camera sensor, comprising: The scanning camera sensor acquires an image of a scene illuminated by a source emitting periodic light pulses based on the integration time, such that when the integration time is different from the period of the light pulses, the image has regular continuous stripes with different brightness. The processor generates feature vectors representing regular continuous stripes with different brightness for the color components of the image, the feature vectors being independent of the reflectivity of objects in the scene and the light level in the scene; The processor determines the frequency of the bands in the image based on the generated feature vector; The processor determines the period of the pulse from the source based on the frequency determined by the stripe in the image; as well as The integration time of the camera sensor is adjusted by the controller so that the integration time is a multiple of the period of the light pulse of the source; The feature vector is generated as a feature vector associated with the brightness of the image, and For rows of the image, the method further includes: The processor obtains the representation value of each color component of the row by summing the values of pixels with the same color component in the row. The processor calculates the value representing the brightness of the row based on the values of each color component representing the same row. The processor calculates the logarithm of the value representing the brightness of the row, and The processor calculates the difference between the logarithm of the value representing the brightness of the row and the logarithm of the value representing the brightness of the adjacent row of the image to generate the feature vector for each of the multiple color components of the image.
13. The method of claim 12, wherein the processor calculates the value representing the brightness of the row based on the values of the respective color components representing the same row and adjacent rows.
14. A method for adjusting the integration time of a scanning camera sensor, comprising: The scanning camera sensor acquires an image of a scene illuminated by a source emitting periodic light pulses based on the integration time, such that when the integration time is different from the period of the light pulses, the image has regular continuous stripes with different brightness. The processor generates feature vectors representing regular continuous stripes with different brightness for the color components of the image, the feature vectors being independent of the reflectivity of objects in the scene and the light level in the scene; The processor determines the frequency of the bands in the image based on the generated feature vector; The processor determines the period of the pulse from the source based on the frequency determined by the stripe in the image; as well as The integration time of the camera sensor is adjusted by the controller so that the integration time is a multiple of the period of the light pulse of the source; The feature vector is generated as a feature vector associated with the color components of the image, and For rows of the image, the method further includes: The processor sums the values of the pixels in the row that have the color components in a manner that represents the values of the color components of the row, and The processor divides the value representing the color component by the value representing the same color component in the same row of a subsequent or previous image in a series of images to generate the feature vector for each of the multiple color components of the image.
15. A method for adjusting the integration time of a scanning camera sensor, comprising: The scanning camera sensor acquires an image of a scene illuminated by a source emitting periodic light pulses based on the integration time, such that when the integration time is different from the period of the light pulses, the image has regular continuous stripes with different brightness. The processor generates feature vectors representing regular continuous stripes with different brightness for the color components of the image, the feature vectors being independent of the reflectivity of objects in the scene and the light level in the scene; The processor determines the frequency of the bands in the image based on the generated feature vector; The processor determines the period of the pulse from the source based on the frequency determined by the stripe in the image; as well as The integration time of the camera sensor is adjusted by the controller so that the integration time is a multiple of the period of the light pulse of the source; The feature vector is generated as a feature vector associated with the color components of the image, and For rows of the image, the method further includes: The processor sums the values of the pixels in the row that have the color components in a manner that represents the values of the color components in the row. Calculate the logarithm of the value representing the color component of the row, and The difference between the logarithm of the value of the color component representing the row and the logarithm of the value of the same color component representing the same row in a subsequent or previous image in a series of images is calculated to generate the feature vector for each of the multiple color components of the image.
16. A method for adjusting the integration time of a scanning camera sensor, comprising: The scanning camera sensor acquires an image of a scene illuminated by a source emitting periodic light pulses based on the integration time, such that when the integration time is different from the period of the light pulses, the image has regular continuous stripes with different brightness. The processor generates feature vectors representing regular continuous stripes with different brightness for the color components of the image, the feature vectors being independent of the reflectivity of objects in the scene and the light level in the scene; The processor determines the frequency of the bands in the image based on the generated feature vector; The processor determines the period of the pulse from the source based on the frequency determined by the stripe in the image; as well as The integration time of the camera sensor is adjusted by the controller so that the integration time is a multiple of the period of the light pulse of the source; The feature vector is generated as a feature vector associated with the brightness of the image, and For rows of the image, the method further includes: The processor obtains the representation value of each color component of the row by summing the values of pixels with the same color component in the row. The processor calculates the brightness value representing the row based on the values of each color component representing the same row and adjacent rows. The processor calculates the logarithm of the value representing the brightness of the row, and The processor calculates the difference between the logarithm of the value representing the brightness of the row and the logarithm of the value representing the brightness of the same row in a subsequent or previous image in a series of images, to generate the feature vector for each of the multiple color components of the image.