FR3151958A1 - METHOD FOR VISUALIZING A LASER SPOT IN A CORRECTED COLOR IMAGE AND IMAGE DETECTION DEVICE IMPLEMENTING THE METHOD - Google Patents

Abstract

The invention proposes an image detection device intended to visualize a laser spot on a color image. The device comprises a matrix of photosites and a matrix of microfilters comprising four types of distinct microfilters, of which three first types have a transmission in two spectral domains, and a fourth type has a transmission in only one of the two spectral domains. The device further comprises an optical band-stop filter in a subdomain of a second spectral domain. The device is configured to construct (301) an image for each type of microfilter; correct (302) each image resulting from a first type of microfilter, by subtracting values, weighted by a coefficient, of pixels from the image resulting from the fourth type of microfilter; obtain (303) a color image from the corrected images, detect (304) a laser spot in the image resulting from the fourth type of microfilter and insert (305) the laser spot into the color image. Figure to be published with the abstract: Fig. 3

Description

METHOD FOR VISUALIZING A LASER SPOT IN A CORRECTED COLOR IMAGE AND IMAGE DETECTION DEVICE IMPLEMENTING THE METHOD

The present invention relates to the field of image sensors and more particularly relates to a method intended to visualize a color image of a scene as well as a laser spot generated by an infrared laser pointer and present in said scene.

STATE OF PRIOR ART

Image sensors of the CMOS silicon type have the advantage of being manufactured in large numbers and at low cost. They have recognized applications in the detection of color images, thanks to their detection sensitivity in the visible spectral range, located at wavelengths ranging from 400 nm to 780 nm. To obtain a color image, an array of colored microfilters, such as a Bayer array, is usually placed in front of an array of photosites of the image sensor. The array of microfilters generally includes three types of colored filters, for example red, green and blue filters. A color image can then be constructed by taking into account, for each pixel of the image, signal levels received by photosites through the different types of colored filters.

In order to obtain a color image with good colorimetric fidelity, in other words an image whose apparent colors are as close as possible to the colors of the imaged scene as perceived by the human eye, it is necessary to isolate signals from radiation with a wavelength located in the visible range from signals from radiation with a wavelength located outside the visible range, for example by avoiding the detection, by the photosites, of infrared radiation. However, the spectral detection range of a CMOS sensor on silicon extends, due to the intrinsic bandgap energy of silicon, to the near infrared, at wavelengths beyond 780 nm. In addition, the filters of the microfilter array generally have a transmission profile that also extends to the near infrared, due to design constraints of the microfilters. Colorimetric fidelity can then be increased by using an additional low-pass filter, making it possible to stop or cut radiation in the near infrared, with a wavelength greater than 780 nm.

In some contexts, and in particular in a military application context, it is interesting to visualize a laser spot generated by an infrared laser pointer and present in a scene, for example for stealth reasons. The detection of such a laser spot requires the detection of optical radiation located in an infrared spectral range, for example in a spectral range located between 800 and 1000 nm in wavelength. The use of such a low-pass filter, making it possible to stop radiation in the near infrared, therefore prevents the detection of the laser spot.

Document FR 3 039 290 describes a method for visualizing a laser spot, in the infrared range, requiring detections of a scene in several spectral bands exhibiting spectral continuity. However, the method described does not make it possible to obtain a color image or to improve its colorimetric fidelity.

It is then desirable to overcome these drawbacks of the state of the art.

It is particularly desirable to provide a solution that makes it possible to visualize a laser spot, in the infrared domain, while making it possible to obtain a color image and improving the colorimetric fidelity of said color image with respect to the imaged scene. It is furthermore desirable to provide a low-cost solution implemented by a single image sensor. Finally, it is desirable to provide a solution that avoids a spatial offset between the image of the laser spot and the image of the scene and that more particularly avoids the introduction of an alignment bias between the image of the laser spot and the image of the scene.

An object of the present invention is to provide an image detection device intended to visualize a laser spot on a color image, the image detection device being of the CMOS sensor type, comprising a matrix of photosites, comprising a matrix of microfilters arranged so that each photosite is arranged facing a microfilter of the matrix of microfilters, and comprising a lens configured to produce an image of a scene on the matrix of photosites. The matrix of microfilters comprises at least four types of microfilters each having a different spectral transmission range, at least three first types of microfilters each having a spectral transmission range included in a first spectral domain and in a second spectral domain, the spectral transmission range of a fourth type of microfilter being included only in the second spectral domain. The image detection device further comprises an optical band-stop filter having a spectral extinction range extending in a first spectral subdomain of the second spectral domain, and comprises a processing unit comprising electronic circuitry configured to: construct an image associated with each type of microfilter and having a predetermined number of pixels, each pixel of the image comprising a value representative of a signal level detected by a photosite associated with said type of microfilter or interpolated from signal levels detected by photosites associated with said type of microfilter; correct each image associated with one of the at least three first types of microfilters, by subtracting, from the value of each pixel of a determined position in said image, the value of the pixel at said determined position of the image associated with the filter of the fourth type, said value being weighted by a coefficient specific to the type of microfilter associated with each corrected image; obtain a color image from all of the corrected images each associated with one of the at least three first types of microfilters; detecting a laser spot having a predetermined shape in the image associated with the fourth type of microfilter and inserting the laser spot into the resulting color image.

According to a particular embodiment, the image detection device further comprises an optical filtering system having a first transmission rate in the first spectral domain and a second transmission rate, distinct from the first transmission rate, in the second spectral domain.

According to a particular embodiment, the image detection device further comprises an optical filtering system having a variable and adjustable transmission spectral width in the second spectral domain, said optical filtering system being configured to cut optical radiation received outside of said transmission spectral width in the second spectral domain.

According to a particular embodiment, the device is sized to obtain a focusing task, in the plane of the photosite matrix, of dimension greater than the dimension of at least two photosites.

The invention also relates to a method for visualizing a laser spot on a color image, the method being implemented by an image detection device, of the CMOS sensor type, comprising a matrix of photosites, comprising a matrix of microfilters arranged so that each photosite is arranged facing a microfilter of the matrix of microfilters, and comprising a lens configured to produce an image of a scene on the matrix of photosites. The matrix of microfilters comprises at least four types of microfilters each having a different spectral transmission range, at least three first types of microfilters each having a spectral transmission range included in a first spectral domain and in a second spectral domain, the spectral transmission range of a fourth type of microfilter being included only in the second spectral domain. The image detection device further comprises an optical band-stop filter having a spectral extinction range extending in a first spectral subdomain of the second spectral domain, and comprises steps implemented by a processing unit of the image detection device, of: constructing an image associated with each type of microfilter and having a predetermined number of pixels, each pixel of the image comprising a value representative of a signal level detected by a photosite associated with said type of microfilter or interpolated from signal levels detected by photosites associated with said type of microfilter; correcting each image associated with one of the at least three first types of microfilters, by subtracting, from the value of each pixel of a determined position in said image, the value of the pixel at said determined position of the image associated with the filter of the fourth type, said value being weighted by a coefficient specific to the type of microfilter associated with each corrected image; obtaining a color image from all of the corrected images each associated with one of the at least three first types of microfilters; detecting a laser spot having a predetermined shape in the image associated with the fourth type of microfilter and inserting the laser spot into the resulting color image.

The above-mentioned and other features of the invention will become more apparent from the following description of at least one exemplary embodiment, said description being given in relation to the accompanying drawings, among which:

schematically illustrates an image detection device;

schematically illustrates an example of a microfilter array of the image sensing device;

schematically illustrates a method of visualizing a laser spot in a corrected color image; and

schematically illustrates an example of a hardware platform suitable for implementing a processing unit of the image detection device.

DETAILED PRESENTATION OF IMPLEMENTATION METHODS

There thus schematically illustrates an image detection device 1. The image detection device 1 is a CMOS type sensor and comprises a matrix of photosites 101. Each photosite is a unit for detecting the received optical radiation and is configured to transform said optical radiation into an electrical signal. The photosite thus detects a signal level representative of a light intensity and converts it into electrical form in order to transmit it for example to a processing unit 105 (described below).

The image detection device 1 further comprises a microfilter array 102 arranged such that each microfilter is arranged facing a photosite of the photosite array 101. According to one embodiment, each microfilter is arranged facing a single photosite. Alternatively, each microfilter is arranged facing a plurality of photosites, for example facing two photosites. The microfilter array 102 is arranged in front of the photosite array. The term in front is here defined relative to a direction 110 of incidence of the optical radiation detected by the photosites. Preferably, the microfilter array 102 comprises as many microfilters as there are photosites in the photosite array 101 and each photosite of the photosite array 101 is associated with a microfilter.

The microfilter array 102 comprises at least four types of microfilters each having a different transmission spectral range. A different transmission spectral range means that transmission levels in at least a portion of the transmission spectral range are different between two types of microfilters. A transmission spectral range of a filter element is defined as the spectral range at which the transmission level is greater than a predefined transmission threshold, or the spectral range at which the normalized transmission level, relative to a maximum transmission level obtained by the filter, is greater than a predefined percentage, for example greater than 20%. A transmission spectral range is not necessarily continuous and may thus comprise several disjoint spectral bands in which the transmission level is greater than the predefined transmission threshold.

At least three first types of microfilters of the microfilter array 102 have a spectral transmission range in the visible range and in the near infrared range. The visible range is defined by a wavelength range between 400 nm and 780 nm. The near infrared range is defined by a wavelength range between 780 nm and 1000 nm. In the visible range, the spectral transmission range is different for each of the first three types of microfilters so as to allow detection of optical radiation of distinct colors.

A fourth type of microfilters of the microfilter array 102 has a transmission spectral range comprised only in the near infrared region.

There schematically illustrates an exemplary microfilter array 102, comprising 16 microfilters distributed in a periodic pattern of 4x4 microfilters.

According to this example, the microfilter array 102 comprises four types of microfilters. The at least three first types of microfilters comprise a first type of microfilter, called a red microfilter R, a second type of microfilter, called a green microfilter V, and a third type of microfilter, called a blue microfilter B. The fourth type of microfilter is an infrared microfilter IR.

In the visible range, the spectral transmission range of the red microfilter R extends for example from 580 nm to 700 nm wavelength, the spectral transmission range of the green microfilter V extends for example from 500 nm to 600 nm wavelength, and the spectral transmission range of the blue microfilter B extends for example from 400 nm to 500 nm. The first types of red microfilters R, green microfilters V and blue microfilters B are presented here as examples, but microfilters having other spectral transmission ranges can be used alternatively, or in addition, such as cyan, magenta or yellow microfilters.

In the near infrared range, each of the first three or more types of microfilters has a distinct transmission profile. A transmission profile is defined by the transmission level of the microfilter as a function of wavelength.

The near-infrared range is divided into a first near-infrared spectral sub-range, for example between 780 nm and 800 nm in wavelength, and a second near-infrared spectral sub-range, for example between 800 nm and 1000 nm in wavelength. In the first near-infrared spectral sub-range, the transmission profiles of the at least three first types of filters differ from one another. On the other hand, in the second near-infrared spectral sub-range, the transmission profiles of said at least three first types of filters are proportional to one another.

The fourth type of microfilter, called IR infrared microfilter, of the microfilter array 102 has a transmission lower than the predefined transmission threshold in the visible spectral range and thus has a spectral transmission range included only in the near infrared range. The IR infrared microfilters have a transmission profile proportional to the transmission profiles of the red R, green V and blue B microfilters in the second infrared spectral sub-range. The transmission profile of the IR infrared microfilters may be different from the transmission profiles of the red R, green V and blue V microfilters in the first infrared spectral sub-range.

The filters of each of the microfilter types are distributed regularly on the microfilter matrix 102. According to the example illustrated in , the microfilter array 102 comprises eight green V microfilters, two blue B microfilters, two red R microfilters and four infrared IR microfilters.

Back to the , the image detection device 1 further comprises a lens 103 for producing an image of the scene observed on the photosite array 101. The lens is located in front of the microfilter array 102. Furthermore, the microfilter array 102 is arranged between the lens 103 and the photosite array 101 at a position such that all the optical radiation detected by each photosite is transmitted by the microfilter located opposite said photosite. The microfilter array 102 is preferably deposited on the photosite array 101, in other words in contact with the photosite array 101.

According to one embodiment, the image detection device 1 is sized to obtain a focusing task, in the plane of the photosite matrix 101, of a dimension greater than the dimension of a photosite, and preferentially covering a plurality of photosites, for example covering 2 to 4 photosites. Thus, each optical signal is received through several types of microfilters. Spectrum folding effects resulting from the presence of different types of microfilters facing contiguous photosites can then be avoided.

The image detection device 1 further comprises an optical band-stop filter 104 having a spectral extinction range extending at least to the first spectral subdomain of the near infrared, located for example between 780 nm and 800 nm in wavelength. A spectral extinction range is defined as a spectral range at which the transmission level of the filter is lower than the predefined transmission threshold. According to one embodiment, the spectral extinction range of the optical notch filter 104 extends below the first near-infrared spectral sub-domain, for example between 700 nm and 800 nm, but not below the highest transmission wavelength, in the visible, of the first types of microfilters, for example not below 700 nm when using red R microfilters. According to one embodiment, the spectral extinction range of the optical notch filter 104 may extend beyond the first near-infrared spectral sub-domain, for example beyond 800 nm, but may not extend up to the highest wavelength of the second near-infrared spectral sub-domain, for example 1000 nm, such that at least a portion of the radiation received in the second near-infrared spectral sub-domain can be detected by the photosites.

The optical band-stop filter 104 is arranged so as to face all of the photosites of the photosite array 101. The optical band-stop filter 104 is integrated into the image detection device 1 and located between the imaged scene and the photosite array 101. According to one embodiment, the optical band-stop filter 104 is arranged in front of the lens 103. According to one embodiment, the optical band-stop filter 104 is integrated into the lens 103, for example by choosing a suitable material or by means of a surface treatment. The optical band-stop filter 104 thus makes it possible to prevent any radiation of a wavelength belonging to the first spectral subdomain of the near infrared from being detected by the photosites of the photosite array 101.

The image detection device 1 comprises the processing unit 105 configured to receive, for each photosite of the photosite matrix 101, information representative of a signal level detected by the photosite in question. Each information representative of a signal level is received in association with a position of the photosite in the photosite matrix 101 and is further associated with a type of microfilter of the microfilter matrix 102, using for example a correspondence table, recorded in the memory of the processing unit 105 and associating each position of the photosite matrix with a type of microfilter. The position of a photosite is for example defined by coordinates such as a row number and a column number.

The processing unit 105 is further configured to process the information received and representative of signal levels and to construct an image from said signal levels. In general, the processing unit comprises electronic circuitry configured to implement the steps of the image detection method described in .

According to one embodiment, the image detection device 1 further comprises a filtering optical system 106. The filtering optical system 106 has a first transmission level in the visible range, between 400 nm and 780 nm wavelength for example, and a second transmission level, distinct from the first transmission level, in the second near-infrared spectral sub-range, between 800 nm and 1000 nm wavelength for example. The first transmission level and the second transmission level are determined so as to compensate for a difference in average light intensity received in the visible range on the one hand and in the second near-infrared spectral sub-range on the other hand. The relative difference between the first transmission level and the second transmission level is for example determined prior to viewing a scene by taking into account estimated brightness parameters associated with the observation context of the scene on the one hand, and on the other hand with an emission intensity of an infrared laser pointer generating a laser spot. The optical filtering system 106 thus makes it possible to detect a laser spot simultaneously with a scene in which the laser spot is located even when the light intensity emitted by the laser spot is low compared to the detected light intensity emitted by the scene or vice versa.

According to one embodiment, the optical filtering system 106 comprises one or more additional optical filters for adjusting the first transmission level relative to the second transmission level. The optical filtering system 106 is integrated into the image detection device 1 and located between the imaged scene and the photosite matrix 101. According to one embodiment, the optical filtering system 106 is arranged in front of the band-stop filter 104, as illustrated in , but may alternatively be arranged in front of the lens 103, between the lens 103 and the microfilter array 102, or integrated into the lens by a surface treatment for example. According to another embodiment, the optical filtering system 106 is included in the optical band-stop filter 104, which comprises distinct transmission levels in the visible range and in the second spectral sub-range of the near infrared.

Alternatively, the first and second transmission levels are equivalent but the transmission spectral width in the second near-infrared spectral subdomain is variable and can be widened or narrowed so as to compensate for a difference in average light intensity received in the visible range on the one hand and in the second near-infrared spectral subdomain on the other hand. According to one example, the optical filtering system 106 comprises a first high-pass interference filter, with a first cut-off wavelength located in the second near-infrared spectral subdomain, and a second low-pass interference filter, with a second cut-off wavelength located in the second near-infrared spectral subdomain and greater than the first cut-off wavelength. The transmission spectral width in said second spectral subdomain is therefore delimited by the first and second cut-off wavelengths. It is then possible to vary said transmission spectral width in a determined manner by tilting the first interference filter and/or the second interference filter by a determined angle. Said transmission spectral width then makes it possible to compensate for a difference in average light intensity received in the visible range on the one hand and in the second near-infrared spectral sub-range on the other hand.

There schematically illustrates a method for visualizing a laser spot in a corrected color image. The visualization method is implemented by the image detection device 1 and comprises processing steps, implemented by the processing unit 105, from the information representative of signal levels detected by the photosites of the photosite matrix 101 and received from said photosite matrix in electrical form. Each information representative of a signal level detected by a photosite is associated with a position of the photosite in the photosite matrix 101 and with a type of microfilter through which said signal is detected.

In a first step 301, the processing unit 105 constructs a first image associated with each type of microfilter. Each constructed image comprises a predetermined number of pixels, such that all the first images comprise the same number of pixels.

The processing unit 105 determines, for each item of information representative of a signal level detected by a photosite, a pixel value representative for example of an intensity level to be displayed on a screen. Each pixel value is thus proportional to the signal level detected by a photosite, and is further associated with the position of said photosite and the type of microfilter through which the signal was received.

The processing unit 105 performs, for each type of microfilter, a reconstruction operation, conventionally called dematrixing, so that each first image comprises as many pixel values as there are pixels, and preferably as many pixel values as there are photosites in the photosite matrix 101.

For example, for each type of microfilter, the processing unit 105 determines the positions, in the photosite matrix 101, for which a pixel value associated with said type of microfilter is determined from a detected signal level. For each position of the photosite matrix 101 for which no pixel value associated with said type of microfilter is determined, the processing unit 105 then determines a pixel value called an interpolated pixel value, by interpolation of determined pixel values, in other words representative of signal levels actually detected, and associated with said type of microfilter. For example, an interpolated pixel value is obtained, for a type of microfilter and at a determined position, from the pixel values associated with the n positions closest to the determined position, n being an integer. The interpolation of the pixel values may be linear or non-linear.

According to one embodiment, the processing unit 105 performs the reconstruction operation from the received information representative of detected signal levels and then converts each detected signal level and each interpolated signal level into a pixel value.

In a following step 302, the processing unit 105 corrects each image associated with one of the at least three first types of microfilters in order to limit the effect produced by the detection of a component of optical radiation located outside the visible range, by the photosites.

Since the image detection device 1 comprises the band-stop filter 104 having a spectral extinction range in the first near-infrared sub-domain, the component of the optical radiation detected in said first sub-domain is therefore zero or negligible compared to the total optical radiation received.

In the absence of the band-stop filter 104, the signal detected by a photosite associated with a first type i of microfilter, i being an integer between 1 and the total number of first types of microfilters, comprises a radiation component (or signal) in the visible range and a radiation component in the near infrared range and can thus be written:

The radiation component in the near-infrared spectral range is the combination of the radiation components And respectively in the first near-infrared spectral subdomain and in the second near-infrared spectral subdomain:

Thus, the presence of the band-stop filter 104 which stops the radiation component in the first near-infrared spectral subdomain and therefore prevents it from being received by each photosite results in detection, by a photosite associated with a first type i of microfilter, of radiation such as :

It is then necessary, to obtain only the radiation component in the visible and thus increase the colorimetric fidelity of the image, to subtract, from the optical radiation detected by a photosite associated with a first type i of microfilter, the radiation component in the second spectral subdomain of the infrared:

Now, the radiation component in the second near-infrared spectral subdomain received by each first type i of microfilter is considered to be proportional to the optical radiation component in the second near-infrared spectral subdomain received by the fourth type of microfilter and denoted :

being a proportionality factor specific to the first type i of microfilter in question.

Furthermore, the radiation detected by a photosite associated with the fourth type of microfilter does not include any radiation component in the first near-infrared spectral subdomain due to the presence of the band-stop filter 104, nor any radiation component in the visible, due to the intrinsic transmission spectral range of the fourth type of microfilter. The radiation detected by a photosite associated with the fourth type of microfilter therefore only includes the optical radiation component in the second spectral subdomain of the near infrared is written:

Thus, to limit the effect of the radiation component in the second spectral subdomain of the near infrared received by a photosite associated with a first type i of microfilter, it is possible to subtract, from the optical radiation detected by said photosite, the radiation detected by a photosite with the same coordinates in the photosite matrix, and associated with the fourth type of microfilter:

To carry out a correction aimed at limiting the effect produced on a color image by the detection of a radiation component located outside the visible range, the processing unit 105 therefore subtracts, pixel by pixel, from the value of the pixel of each first image associated with one of the at least three first types of microfilters, a value obtained by weighting, by the coefficient specific to the first type i of microfilter associated with the corrected image, of the value of the pixel of the image associated with the fourth type filter. The processing unit 105 performs the subtraction for each pixel defined by its position (i.e. its coordinates) in the image.

The coefficient specific to each type i of microfilter is determined prior to the implementation of the visualization method, for example during a calibration step, and recorded in the memory of the processing unit 105. The coefficients are independent of the observed scene. Each coefficient is defined as the coefficient of proportionality, in the second near-infrared spectral subdomain, between a photosite detection profile associated with one of the at least three first types of microfilters and a photosite detection profile associated with the fourth type of microfilter. It is thus possible to obtain pixel values, for each of the at least three first types of microfilters, that are as faithful as possible to the optical radiation received, in the visible range, by the photosites and thus to increase the colorimetric fidelity of a color image relative to the imaged scene.

In a following step 303, the processing unit 105 constructs a color image from all of the corrected images each associated with one of the at least three first types of microfilters. For example, for a microfilter matrix 102 comprising red R, green G and blue B microfilters, the processing unit 105 determines for each pixel, identified by its position in an image, a triplet of pixel values comprising the values of said pixel of each image associated with the red R, green G and blue B microfilters. Each triplet of pixel values corresponds to a predefined color. The processing unit 105 can then transmit to a display device the predefined color corresponding to each pixel. The processing unit 105 can further perform colorimetric correction and/or white balance operations.

In a following step 304, the processing unit 105 detects in the first image associated with the fourth type of microfilter, in other words associated with the infrared microfilter IR, a laser spot from an infrared laser pointer. The spectral emission range of the infrared laser pointer is included in the second spectral subdomain of the near infrared, which makes it possible to ensure that the laser spot is present in said first image associated with the infrared microfilter IR. The detection of the laser spot is carried out by detecting a predetermined shape, in a determined spectral range, for example by implementing at least one image processing operation of the thresholding, local contrast expansion and frequency filtering operations. The detection of the laser spot from an infrared laser pointer differs from a detection implemented using a conventional decamouflage algorithm which searches, among a set of different spectral bands, for spectral singularities of undetermined shapes. According to the method of the present invention, the detection of the laser spot from an infrared laser pointer involves, on the other hand, searching for a predetermined shape in a determined spectral band which corresponds to the second spectral subdomain of the near infrared. The predetermined shape, which corresponds to the shape of the laser spot, is for example a Gaussian, pseudo-Gaussian, rotationally symmetrical beam shape, or a shape without rotational symmetry when the laser beam is shaped by beam-shaping techniques.

According to one embodiment, the optical filtering system 106 as described in relation to the is inserted into the image detection device 1 in order to adjust the average light intensity received from the laser spot relative to the average intensity received from the scene. Indeed, the detection of the laser spot can be limited when the light intensity emitted by the laser spot is low relative to the light intensity emitted by the scene, for example due to a saturation of detection of the photosites, associated with the infrared microfilters IR, by the light intensity received from the scene, or due to a light intensity received from the laser spot too low to be distinguished from a noise level of said photosites. The insertion of the optical filtering system 106 makes it possible to overcome these limitations of detection of the laser spot, with for example the first transmission level, in the visible range, lower than the second transmission level, in the second near-infrared sub-range.

Conversely, when the light intensity emitted by the laser spot is too high relative to the light intensity emitted by the scene, the light intensity received by the photosites associated with the first types of microfilters from the scene may be too low relative to a noise level of said photosites, or said photosites may be saturated in an area corresponding to the laser spot, thus preventing at least part of the scene from being viewed. Inserting the optical filtering system 106 makes it possible to overcome this limitation of viewing the scene, for example with the first transmission level, in the visible range, higher than the second transmission level, in the second near-infrared sub-range, or with a transmission spectral width in the second near-infrared sub-range narrowed relative to the total width of the second near-infrared spectral sub-range.

In a following step 305, the processing unit 105 inserts the laser spot into the constructed color image. The insertion of the laser spot is carried out for example by replacing pixels of the color image with pixels of the same position of the image associated with the fourth type of microfilter and comprising the detected laser spot. According to another example, said pixels of the color image and of the image associated with the fourth type of microfilter are mixed.

It is thus possible to obtain an image of a scene having good colorimetric fidelity and comprising an image of a laser spot originating from an infrared laser pointer and observable in the scene. Furthermore, the image of the laser spot does not have any spatial offset relative to the image of the scene since the laser spot and the scene are both imaged by a single image detection device 1, in other words by a single sensor. A pixel coordinate in the image of the scene and in the image of the laser spot therefore corresponds to a single direction in the object space. No alignment bias between the image of the spot and the image of the scene is therefore introduced thanks to the visualization method of the present invention.

There schematically illustrates an example of a hardware platform for implementing, in the form of electronic circuitry, the processing unit 105.

The hardware platform comprises, connected by a communication bus 410: a processor or CPU (“Central Processing Unit” in English) 401; a RAM (“Random-Access Memory” in English) 402; a read-only memory 403, for example of the ROM (“Read Only Memory” in English) or EEPROM (“Electrically-Erasable Programmable ROM” in English) type, such as a Flash memory; a storage unit, such as a hard disk HDD (“Hard Disk Drive” in English) 404, or a storage media reader, such as an SD (“Secure Digital” in English) card reader; and an I/f interface manager 405.

The I/f interface manager 405 allows the processing unit 105 to interact with the photosite array 101, as well as with a display device if applicable.

The processor 401 is capable of executing instructions loaded into the RAM 402 from the ROM 403, an external memory, a storage medium (such as an SD card), or a communications network. When the hardware platform is powered on, the processor 401 is capable of reading instructions from the RAM 402 and executing them. These instructions form a computer program causing the processor 401 to implement some or all of the steps and methods described herein.

All or part of the steps, methods and operations described herein may thus be implemented in software form by executing a set of instructions by a programmable machine, for example a DSP (Digital Signal Processor) type processor or a microcontroller, or be implemented in hardware form by a machine or a dedicated electronic component (chip) or a set of dedicated electronic components (chipset), for example an FPGA (Field Programmable Gate Array) or ASIC (Application Specific Integrated Circuit) component. Generally speaking, the processing unit 105 comprises electronic circuitry adapted and configured to implement the methods and steps described herein.

Claims (5)

Image detection device (1) intended to visualize a laser spot on a color image, the image detection device being of the CMOS sensor type, comprising a matrix of photosites (101), comprising a matrix of microfilters (102) arranged so that each photosite is arranged facing a microfilter of the matrix of microfilters (102) and comprising an objective (103) configured to produce an image of a scene on the matrix of photosites (101),
the image detection device (1) being characterized in that the microfilter matrix (102) comprises at least four types of microfilters (R, G, B, IR) each having a different spectral transmission range, at least three first types of microfilters (R, G, B) each having a spectral transmission range included in a first spectral domain and in a second spectral domain, the spectral transmission range of a fourth type of microfilter (IR) being included only in the second spectral domain,
in that the image detection device (1) further comprises an optical band-stop filter (104) having a spectral extinction range extending into a first spectral subdomain of the second spectral domain, and in that it comprises a processing unit (105) comprising electronic circuitry configured to:
- constructing (301) an image associated with each type of microfilter and having a predetermined number of pixels, each pixel of the image comprising a value representative of a signal level detected by a photosite associated with said type of microfilter or interpolated from signal levels detected by photosites associated with said type of microfilter,
- correcting (302) each image associated with one of the at least three first types of microfilters, by subtracting, from the value of each pixel of determined position in said image, the value of the pixel at said determined position of the image associated with the filter of the fourth type, said value being weighted by a coefficient specific to the type of microfilter associated with each corrected image,
- obtaining (303) a color image from all the corrected images each associated with one of the at least three first types of microfilters,
- detecting (304) a laser spot having a predetermined shape in the image associated with the fourth type of microfilter and inserting (305) the laser spot into the color image obtained. An image sensing device according to claim 1, further comprising an optical filter system (106) having a first transmission rate in the first spectral domain and a second transmission rate, distinct from the first transmission rate, in the second spectral domain. An image sensing device according to claim 1, further comprising an optical filter system (106) having a variable and adjustable transmission spectral width in the second spectral domain, said optical filter system (6) being configured to cut received optical radiation outside said transmission spectral width in the second spectral domain. Image detection device according to one of claims 1 to 3, the device being sized to obtain a focusing task, in the plane of the photosite matrix (101), of dimension greater than the dimension of at least two photosites. Method for visualizing a laser spot on a color image, the method being implemented by an image detection device (1), of the CMOS sensor type, comprising a matrix of photosites (101), comprising a matrix of microfilters (102) arranged so that each photosite is arranged facing a microfilter of the matrix of microfilters, and comprising an objective (103) configured to produce an image of a scene on the matrix of photosites (101),
the method being characterized in that the microfilter matrix comprises at least four types of microfilters each having a different spectral transmission range, at least three first types of microfilters each having a spectral transmission range included in a first spectral domain and in a second spectral domain, the spectral transmission range of a fourth type of microfilter being included only in the second spectral domain,
in that the image detection device further comprises an optical band-stop filter having a spectral extinction range extending into a first spectral subdomain of the second spectral domain, and in that it comprises steps implemented by a processing unit of the image detection device, of:
- constructing (301) an image associated with each type of microfilter and having a predetermined number of pixels, each pixel of the image comprising a value representative of a signal level detected by a photosite associated with said type of microfilter or interpolated from signal levels detected by photosites associated with said type of microfilter,
- correcting (302) each image associated with one of the at least three first types of microfilters, by subtracting, from the value of each pixel of determined position in said image, the value of the pixel at said determined position of the image associated with the filter of the fourth type, said value being weighted by a coefficient specific to the type of microfilter associated with each corrected image,
- obtaining (303) a color image from all the corrected images each associated with one of the at least three first types of microfilters,
- detecting (304) a laser spot having a predetermined shape in the image associated with the fourth type of microfilter and inserting (305) the laser spot into the color image obtained.