WO2025243215A1

WO2025243215A1 - Method for controlling the discharge of a photosensitive element in a photodetector and photodetector thereof

Info

Publication number: WO2025243215A1
Application number: PCT/IB2025/055256
Authority: WO
Inventors: Monica Vatteroni
Original assignee: Eye Tech SRL
Current assignee: Eye Tech SRL
Priority date: 2024-05-21
Filing date: 2025-05-21
Publication date: 2025-11-27
Anticipated expiration: 2026-11-21

Abstract

A method for controlling the discharge of a photosensitive receiving means (23) in a pixel (20), in particular for optical sensors, provides that, at a pixel (20) of the type provided for a transfer gate (26) interposed between the photosensitive receiving means (23) and a read-out node (28), coupled to an element useful for providing a reference and polarization voltage (30), as well as a further transfer gate (21) for the transfer of charge from the photosensitive receiving means (23), the aforesaid further transfer gate (21) is controlled with a variable signal and preferably programmed to obtain a high dynamic of the photosensitive receiving means (23). In a photodetector (10) comprising at least one pixel (20) as outlined above, the further transfer gate (21) is controlled by means of a discharge control circuit (50) to which a charge drainage component (60) is associated.

Description

METHOD FOR CONTROLLING THE DISCHARGE OF A PHOTOSENSITIVE ELEMENT

IN A PHOTODETECTOR AND PHOTODETECTOR THEREOF

FIELD OF APPLICATION

[0001] The present invention relates to a method for controlling the discharge of a photosensitive element in a photodetector, i.e. an electro-optical device, suitable for detecting an incident light and converting it into a related electrical signal.

[0002] The invention also refers to a photodetector device capable of guaranteeing a high image quality both in low brightness conditions and in the presence of high illumination, therefore with variable light intensity over a wide range and also in conditions of high variability of environmental and illumination conditions.

[0003] A photodetector according to the invention is used in particular, although not exclusively, to make electro-optical sensors to be inserted in digital cameras, digital cameras and intelligent vision systems used for various applications such as, but not limited to, autonomous navigation, and in particular autonomous driving, etc.

BACKGROUND OF THE INVENTION

[0004] Electro-optical sensors are known comprising a plurality of photodetector devices suitable for detecting light signals by means of a photosensitive element, and transmitting them, in the form of a charge and/or electrical signal, to a calculation unit that processes them by obtaining images. Such images are then transmitted to display devices or a processing unit, which allow a human or virtual user (such as artificial intelligence algorithms) to display such images or process them in order to extract the information contained therein for the purpose of displaying them or receiving information on the basis of which to make informed decisions.

[0005] Today, such electro-optical sensors are generally based on CMOS (Complementary Metal Oxide Semiconductor) technology with typically silicon substrate, which has an excellent optical behaviour in the visible and near-infrared region, with a light signal acquisition window ranging from 350nm to 1100nm. Other materials are however used today to extend the window of visibility to higher frequencies. These devices typically combine other materials such as, by way of example only, gallium arsenide indium or germanium (EYE4NIR), or bolometers (FLIR thermal cameras) for image detection with electrical signal acquisition and processing circuits based on CMOS technology. [0006] Traditional silicon-based sensors are typically based on a light acquisition mechanism by integrating the light signal onto a photosensitive element capable of accumulating the produced signal. Such sensors are typically characterized by limited light dynamics, i.e. they are able to collect information in a limited light/dark range, mapping as dark (signal corresponding to 0) the signals below the detectable minimum or saturation (signal corresponding to the maximum, 255 on an 8-bit scale) in the case of signal greater than the maximum detectable light. In this type of sensor, the width of the window corresponding to the light dynamics remains constant, while the position of the window can be varied by varying the integration time, otherwise called the exposure time to the light of the sensor, which can be very short, favouring the acquisition of intense lights, or very long, favouring the acquisition of low lights. The light dynamics of these sensors typically range from 45dB to 60dB.

[0007] A very popular implementation of the linear pixel today is the one called 'pinned photodiode'. This technology works with charge transfer and not with electrical signal transfer, guaranteeing better performance, especially in terms of noise.

[0008] The low dynamics of the aforementioned sensors limits their use in many applications where there is a need to acquire information in a wider light range. For example, automotive, where the light can vary from very low (at night or in shady areas) to very high (areas directly illuminated by the sun), even with a high temporal variability.

[0009] To overcome this problem, sensors defined as high light dynamics are typically used, capable of offering high image quality even in the case where images characterized by a high brightness range must be detected.

[0010] Such sensors use different techniques to map information over a very wide range of lights over the available analogue signal range.

[0011] A technique known in the state of the art is that of multi-integration or multi-exposure. This technique involves the acquisition of a plurality of images with different integration/exposure times. The information of the various images is subsequently combined via software to obtain a single image with high light dynamics. This type of technique, based on the acquisition of multiple images, is called inter-frame. The benefit of this technique is that it is conceptually simple and relies on a well known and used linear sensor architecture. The main disadvantages are: the enormous amount of data that must be managed even just to generate a single image, the high noise due to the fact that the information obtained is correlated with each other, secondary effects such as ghosting, the presence of artifacts in the image due to the superimposition of images of moving objects, and flickering, which also affects linear sensors and is manifested when trying to acquire information deriving from LED signals that are typically pulsed, with a frequency that is not perceptible by the human eye but, under certain illumination conditions, is detected by the cameras with a consequent partial or total lack of information in the acquired images. The problem of ghosting is now solved with the simultaneous acquisition of different images, an effective solution that, however, exponentially complicates the complexity of the pixel (ONSEMI, SONY, OMNIVISION), losing the original advantage of architectural simplicity. Today, however, flickering is mitigated through software. The image obtained with the multi-exposure technique results from the mapping of the light on the output signal range with a linear trend in sections.

[0012] A technique widely used today, which tries to address problems such as ghosting, is what is called dual-gain. More images are acquired as in the prior art, varying the pixel gain instead of the integration time. Depending on the architectures implemented, the images with different gain can be acquired sequentially or simultaneously or managed dynamically with the implementation of an appropriate control. This technique is currently the most widely used, although it involves considerable complexity in pixel architecture (ONSemi, SONY, OMNIVISION, US10070081 B2).

[0013] Recently, defined overflow techniques are being used. In this case, the pixel overflow signal is read. In practice, the excess signal above saturation that was typically discarded is now read because it is stored in an additional capacitance/diode. The advantage of this technique is the intraframe read-out of the high-dynamic signal but it also suffers from a high architectural complexity, with the addition of the complexity of managing an additional signal memory, the one that preserves the overflow signal. This technique is also implemented today by the most important operators in the sector (ONSemi, SONY, OMNIVISION, US9917120B2).

[0014] Another technique known in the state of the art is based on a logarithmic scale compression of the signal inside the sensitive element. Solutions are known in which compression is achieved by connecting to the photo-sensitive junction a MOS-type transistor in diode configuration, as described for example in US 5,608,204. The presence of the diode causes part of the charge to be drained with a logarithmic proportion with respect to the incident light, preventing the photosensitive element from saturating. The advantage of this technique is the extreme architectural simplicity, the very high dynamics and the need for a single image acquisition that contains all the necessary information (defined intra-frame). Further, the final image is natively generated by the sensor without the need for processing. Another advantage that particularly distinguishes this technique is the possibility of randomly accessing the various pixels without having to wait for an integration time, a capability that has determined its use in sensors defined as 'Event based' or neuromorphic (PROPHESEE). However, this type of electro-optical sensors are characterized by a high compression of the light signal, a very high residual spatial noise as it is not erasable at the hardware level, and a very slow adaptation to the variation of light in low light conditions that occurs in the form of artifacts, in this case traces in the images when acquiring moving objects, making their content unreadable and difficult to use especially by artificial intelligence algorithms.

[0015] It is also possible to control the exposure time of each photosensitive element individually so as to avoid saturation at the end of the light exposure period. An implementation example of this technique is described in US 6,975,355 and provides for integrating a flip-flop and an AND port in each photosensitive element to control the start time of the integration time and the exposure locally. However, this technique provides for the acquisition of a linear signal and, consequently, although the different photosensitive elements will work with different acquisition intervals and therefore on different dynamics, each photosensitive element individually will have a linear dynamics and therefore a low dynamics when compared to the cases previously mentioned.

[0016] Much of the limitations of these techniques have been overcome with the solutions proposed in the previous patents EP 2761657 and EP 1874699 from the same applicant which present techniques for acquiring high dynamic images based on the control of the discharge of the photosensitive element, appropriately piloting the reset signal of the photosensitive element itself and consequently modeling the electrical signal that maps the power of the incident light (FIGS. 18a and 18b).

[0017] Advantages of this technique are the high flexibility in programming the output light dynamics, both in amplitude and shape, based on the reset control signal and, consequently, the possibility of optimizing the response in terms of dynamics and resolution in the desired light range, rather than increasing its compression and obtaining very high dynamic images. All this is done by acquiring a single image and therefore avoiding artifacts, limiting processing complexity and obtaining high quality in any light condition.

[0018] The limit of the technique presented in the above mentioned patents is the ability to obtain high dynamic images only in case of controlled light. This is because a single reset curve is generated and distributed simultaneously to the entire array of photosensitive elements and therefore it is necessary to interrupt the light flux that affects the pixels to stop the discharge process during the read-out phase. It is clear that this constraint limits the scope of application of the above-mentioned technology to those situations wherein the illumination conditions of the scene can be controlled. This limit was exceeded by the same applicant with patent application EP 4154517, published internationally under no. WO 2021/234566 A1 , which provides for a continuous shift of the reset curve along the pixel array, with a delay effect of propagation of the curve from each line to the next.

[0019] Also in the latter, as in the previous patents cited by the applicant, the generation of the discharge curve applied to the reset signal is provided by means of a signal or a plurality of signals deriving from a digital analogue converter which involves programming it on several digital bits (typically 8) to obtain the signal resolution necessary for correct and accurate operation.

[0020] The limit of the architecture presented in the applicant's cited patents is that they are based on the control of the pixel discharge through the reset voltage. This makes the technique not applicable to next generation pixels (Fig. 18c and Fig. 18d) that physically separate the photosensitive element from the reset transistor during the integration period by interposition of a transfer gate, typically consisting of a transistor in MOS technology with a switch function. Also in the aforementioned application WO 2021/234566 A1 (a configuration of which is shown in Fig. 18b) there is a possible presence of a transfer gate that physically separates and therefore decouples the readout node from the photodiode, however in this case the reset transistor is coupled to the photodiode and not to the read-out node. According to the technique implemented in WO 2021/234566 A1 , the connection between the photosensitive element and the reset signal is in fact necessary in order to implement an effective conditioning in increasing the light dynamics of the photosensitive element. In addition, the connection of the read-out node to the reset signal and the decoupling from the photosensitive element, typical of the new generation pixels, allow the integration of a global shutter that is not possible in the aforementioned architectures of the applicant's patents.

[0021] In next generation pixels (Fig. 18c and Fig. 18d) in which the photosensitive element is decoupled from both the reset signal and the read-out node via a transfer gate, architectures are also known in which there is a further transistor directly connected to the photosensitive element (FIG. 18d) which is used for charge drainage in order to completely empty the photosensitive junction or to implement an anti-blooming effect. In the first case the technique implemented is to provide a pulse before the onset of the integration time that allows complete discharge of the photosensitive junction, while in the second case a constant, non-zero signal is provided during the integration time to prevent the photosensitive junction from overloading. However, none of the known techniques make it possible to increase the dynamics of the photosensitive element.

SUMMARY OF THE INVENTION [0022] One aim of the present invention is to realise a method and structure for controlling pixel discharge that acts directly on the photodetector element by dynamically conditioning its discharge and that can be used in any photosensitive element architecture that implements a photodetector with capacitive behaviour.

[0023] Another general object of the present invention is to realize a photosensitive element that can be integrated in an electro-optical device implemented on a support element, or substrate, suitable for realizing electronic circuits, for example made of silicon, of small dimensions and that is suitable for providing good quality images at a high repetition frequency both in case of low brightness and in the presence of a wide range of brightness present in the observed scene.

[0024] Another object is to provide an electro-optical device that is capable of operating in any light condition, allowing the response dynamics of the sensor to be adapted both to the illumination conditions and to the needs of the user and/or application.

[0025] A further object of the present invention is that the photosensitive elements of the present invention can be arranged in an array, or in another desired arrangement, in an electro-optical sensor that provides for the use of a plurality of photosensitive elements that can be managed independently, or of a desired subset, variable in size and shape, or globally, by controlling the discharge of the photosensitive elements individually or in subgroups.

[0026] Another object of the present invention is to enable the possibility of implementing an electronic shutter on board the sensor which, in addition to ensuring superior image quality, also enables the possibility of using the presented technique in uncontrolled light conditions.

[0027] According to an aspect of the invention, the above objects are achieved by means of a method for controlling the discharge of a photosensitive element according to Claim 1 .

[0028] In accordance with another aspect of the invention, the above objects are achieved by a photodetector according to Claim 4. The dependent claims describe other characteristics of the present invention or variants to the main inventive idea.

[0029] A method according to the present invention comprises at least one circuit for detecting light and converting the detected charge into an electrical signal, said pixel, which in turn comprises:

- at least one photosensitive receiving means, such as a reverse polarized diode,

- a read-out node to which a read-out architecture is connected, the read-out node comprising an element useful for providing the pixel with a reference and polarization voltage, said element being controlled by a reset terminal and being arranged in a configuration decoupled from the photosensitive receiving means,

- a charge or electrical signal transfer gate (depending on the physical implementation of the pixel) controlled by a transfer terminal and interposed between the at least one photosensitive receiving means and the read-out node, so that the read-out node is decoupled from the photosensitive receiving means, and

- a further charge transfer gate from the pixel controlled by a discharge conditioning terminal and coupled directly to the at least one photosensitive receiving means.

[0030] According to the method of the invention, a voltage signal of variable type and in particular of programmable type is applied to the aforementioned discharge conditioning terminal.

[0031] Advantageously, according to the method of the invention the control signal is initially brought to a reference value, and then discharged progressively and in a controlled manner.

[0032] In accordance with another aspect of the invention the above objects are achieved by a photodetector comprising: at least one circuit for detecting light and converting the detected charge into an electrical signal, said pixel, which in turn comprises: at least one photosensitive receiving means, such as a reverse polarized diode, a read-out node to which a read-out architecture is connected, the read-out node comprising an element useful for providing the pixel with a reference and polarization voltage, said element being controlled by a reset terminal and being arranged in a decoupled configuration from the photosensitive receiving means; a charge or electrical signal transfer gate (depending on the physical implementation of the pixel) controlled by a transfer terminal and interposed between the photosensitive receiving means and the read-out node, so that the reset terminal is decoupled from the photosensitive receiving means, and a further charge transfer gate from the pixel coupled directly to the at least one photosensitive receiving means and controlled via a discharge conditioning terminal, a discharge control circuit suitable for bringing a variable voltage to the discharge conditioning terminal which is able to dynamically condition the amount of charge dissipated by the incident light.

[0033] The pixels are typically aggregated in a linear or matrix configuration and share the same read-out line, for this reason each pixel has a switch that allows it to be coupled to or decoupled from the read-out line exclusively with respect to the others. The decoupling of the read-out node from the photosensitive node makes it possible to decouple the read-out phase from the integration phase, making it possible to implement an electric shutter and operate the pixel in uncontrolled light since all the pixels integrate simultaneously.

[0034] In a preferred embodiment a charge drainage component or current sink is associated to the discharge control circuit.

[0035] Advantageously, the discharge control circuit comprises: a charging element which transmits a reference signal and is controlled by a boosting signal. a capacitance capable of maintaining the voltage signal transmitted by the charging element, and a switch, conveniently implemented in CMOS technology with a transistor pass or a gate pass, to selectively connect the charge drainage component.

[0036] The voltage signal is initially brought to a reference value by the charging element and then discharged by the charge drainage component. This discharge takes place in a controlled manner by activating or deactivating the switch. A certain sequence of pulses to the switch makes it possible to create the desired curve and therefore the desired dynamic trend. This curve can have different trends depending on the value of the set reference voltages, the set discharge current and the sequence with which the charge drainage component is connected.

[0037] The proposed structure has many advantages mainly related to the overall dimensions, the precision and flexibility of the curve that can be generated and the simplicity of control that takes place through two signals: reference signal and boosting signal. [0038] The use of a programmable signal makes it possible to detail the brightness range to be acquired in position, amplitude and degree of compression, depending on the type of image processing and application requirements.

[0039] The aforementioned programmable reference signal makes it possible to obtain information related to signals that exceed the linear dynamics according to the known technology with a single acquisition with precision and programmable behaviour. In fact, the programmability of the reference signal guarantees the possibility of obtaining different compressions for various illumination areas of the same scene, depending on the needs related to the type of application and the characteristics of the scene acquired.

[0040] In fact, the light dynamics of the pixel strongly depends on the shape of the signal curve and can vary from linear, like that of a standard pixel, to very high.

[0041] According to a preferred embodiment of the invention, a photodetector device according to the invention comprises two or more pixels having the discharge conditioning terminal in common and connected to a single pixel discharge control circuit.

[0042] According to a preferred embodiment of the invention, in a photodetector device, a single charge drainage component is associated with two or more discharge control voltage generation elements.

[0043] According to yet another aspect of the present invention, the aforementioned objects are achieved by means of an electro-optical sensor comprising a plurality of pixels as defined above, at least one discharge control circuit and at least one charge drainage component, wherein one or more pixels are associated with each discharge control circuit and one or more discharge control circuits are associated with each charge drainage component.

[0044] Advantageously, the electro-optical sensor comprises a plurality of photodetectors and control means, which are connected to the discharge control circuits of the photodetectors and which make it possible to independently or aggregatedly manage the configuration of each of said photodetectors. The possibility of independently controlling the individual photodetectors, either individually or in aggregate, or for independent groups, enables a flexibility that allows the optimization of the response of the electro-optical sensor according to the application needs or the illumination conditions of the scene.

[0045] The arrangement of the individual photodetectors in a sensor is such that the pixels of different photodetector devices form linear or matrix configurations, according to desired arrangements in rows and/or columns. [0046] By combining the photodetector with an appropriate circuit for analyzing the images obtained, it is also possible to create an adaptive circuit in which the response curve of the photosensitive element is optimized in real time as a function of the distribution of the light intensity on the scene in question, the details to be highlighted or any application need. By doing this, the user will always be able to highlight and/or detect all the relevant details of the captured scene.

[0047] The circuit described also has the following advantages over the state of the art: it has the ability to detect light, or light radiation, in a range of intensities that can be varied and modulated as required, up to more than 120 dB, within a single scene, or image, under any image acquisition condition, even in uncontrolled and highly variable light conditions both temporally and spatially; it is managed in a manner analogous to a standard linear sensor well known in the state of the art, facilitating its integration; it provides good image quality even in case of low input brightness and therefore low current generated by the aforementioned photosensitive element; it makes it possible to analyze areas with different brightness of the scene, or of the image, with programmable precision for each photosensitive element or a subset of photosensitive elements; it has the ability to be configured, without interrupting the generated image stream, so as to respond in real time to changes required by any circuits or devices, even based on computing devices or computational blocks, capable of modifying its dynamic response to light based on the current/actual conditions of the observed scene in order to optimize the information content required by the specific application in which the electro-optical device is used.

DESCRIPTION OF THE FIGURES

[0048] These and other features of the present invention will become clear from the following preferred embodiment description, given as a non-limiting example, with reference to the attached drawings wherein:

FIG. 1 illustrates a circuit diagram of a photodetector according to the present invention; FIG. 2 shows a side view of the photodetector of Fig. 1 in which a possible implementation of a charge drainage component of the photodetector is shown;

FIGS. 3 and 4 illustrate two possible variants of the photodetector of Fig.1 , specifically dedicated to systems based on charge transfer such as pinned photodiodes;

FIGS. 5 to 14 illustrate some embodiment variants of electro-optical sensors comprising a plurality of photodetectors according to the invention;

FIGS. 15 and 16 show time diagrams showing the generation mode of two possible curves used for the discharge control according to the method of the invention;

FIG. 17 shows six time diagrams in which the letters a) to f) indicate possible charge control signals applied to a photosensitive element in which diagram a) shows a conventional type signal while diagrams b) to f) show signals according to the method of the present invention;

FIG. 18 shows, indicated with letters a) to d), the wiring diagrams of four different conventional photodetectors.

DISCLOSURE OF SOME PREFERRED EMBODIMENTS

[0049] With reference to Fig. 1 , a photodetector according to the present invention is generally indicated with 10.

[0050] The photodetector 10 comprises a light signal detection and conversion circuit, also called a pixel, 20, a discharge control circuit, 50, and a charge drainage component, or current sink, 60.

[0051] The pixel 20 in turn comprises: at least one photosensitive receiving means, 23; a read-out node, 28, to which a read-out architecture is connected; a transfer gate, 26, controlled by a transfer terminal, 25, said transfer gate 26 being interposed between the at least one photosensitive receiving means 23 and the read-out node, 28, in turn connected to an element useful for providing a reference and polarization voltage 30 so that the reset terminal 27 is decoupled from the photosensitive receiving means 23; a further transfer gate, 21 , for transferring the charge from the photosensitive element 23 coupled directly to the at least one photosensitive receiving means 23, said further transfer gate 21 being controlled via a discharge conditioning terminal, 22.

[0052] In an embodiment of the invention the read-out architecture comprises: a read-out terminal, 35, or pixel output node for connecting the pixel 20 to the external read-out line; a switch, 33, for connecting the rest of the pixel 20 to the read-out terminal 35, said switch 33 being controlled via a selection terminal, 34; an element useful for providing a reference and polarization voltage, 30, to the pixel 20 controlled by a reset terminal, 27; and an active signal amplification and decoupling element, 31 , arranged between the read-out node 28 and the read-out terminal 35.

[0053] The element 30 useful for providing a reference and polarization voltage, controlled by the reset terminal 27 is part of the read-out node 28. Since the read-out node 28 is decoupled from the photosensitive receiving means 23 by means of the interposition of the transfer gate 26, both the read-out terminal 35 and the reset terminal 27 are decoupled from the photosensitive receiving means 23.

[0054] The photosensitive receiving means 23 is a capacitive type element, specifically a light sensitive, inversely polarized diode. In standard CMOS sensors the photosensitive receiving means 23 is integrated into the silicon substrate. In an embodiment in which the sensor is sensitive to wavelengths other than the visible one, the photosensitive receiving means 23 is instead made of materials other than silicon and interfaced with read-out circuits integrated on the silicon.

[0055] The further transfer gate 21 is an element useful for controlling charge dissipation of the photosensitive element 23 and is realized by means of an N-type MOS transistor, a BJT transistor or other FET-type component or assembly of components.

[0056] The transfer gate 26 is also a transistor arranged to transfer the electrical charge or signal from the photosensitive element 23 to a read-out node, 28. [0057] The switch 33, driven by the selection signal 34, carries the output signal from the amplification and decoupling element 31 to the read-out terminal 35.

[0058] According to a preferred embodiment of the invention, the active signal amplification and decoupling element 31 , which is advantageously a transistor configured as a source follower that acts as a buffer, the further transfer gate 21 , the element useful for providing a reference and polarization voltage 30, the transfer gate 26 and the switch 33 are all made of CMOS technology.

[0059] According to the method of the invention, a pixel 20 as described above is controlled by applying to the discharge conditioning terminal 22 a control signal of variable type, preferably programmable.

[0060] According to a preferred embodiment the control signal is initially brought to a reference value, and then discharged progressively and in a controlled manner.

[0061] The method is implemented by providing a discharge control circuit 50, capable of generating the control voltage applied to the discharge conditioning terminal 22, which comprises: a charging element, 51 , which receives a reference signal via a reference terminal, 55, and is controlled by a boosting signal via a boosting terminal, 54 for transmitting a charging signal, a capacitance, 52, capable of maintaining the voltage signal transmitted by the charging element 51 , and a switch, 53, controlled by a synchronization signal, 56, said switch 53 being advantageously implemented in CMOS technology with a transistor pass or a gate pass, for selectively connecting the charge drainage component 60.

[0062] The signal held in the capacitance 52 is conditioned and controlled by the charging element 51 , which makes it possible to bring the capacitance 52 to a known reference value determined by the signals transmitted to the boosting terminal 54 and to the reference terminal 55; and by the synchronization signal transmitted to the switch 53 which makes it possible, by connecting the capacitance 50 to the charge drainage component 60 with a given sequence of pulses, to discharge the capacitance 52 in a controlled manner, generating an appropriate discharge curve, i.e., an appropriate discharge control signal of the photosensitive receiving means 23.

[0063] The charging element 51 and the switch 53 are also made of CMOS technology. The capacitance 52 is also implemented in CMOS technology in ways known in the state of the art. [0064] With reference to Fig. 2, according to an advantageous embodiment the charge drainage component 60 comprises: an active charge transfer element, 62, 63, consisting of two elements in a current mirror configuration;

- A current generation element, 65, which may be replaced with an external signal.

[0065] With reference to FIG. 3, a possible implementation of the pixel 20 of FIG. 1 , operating partially with charge transfer.

[0066] The pixel 20 comprises:

- A photosensitive junction, 37, implementing the photosensitive element 23; a drainage gate, 39, to act on the controlled drainage of the charge; a drainage junction, 36, where the charge drained by the photosensitive junction 37 is transferred; a read-out junction, 38, where charge is transferred from the photosensitive junction 37 for read-out; a read-out gate, 40 adapted to transfer the charge from the photosensitive junction 37 to the read-out junction 38.

[0067] The drainage gate 39 and the drainage junction 36, together with the photosensitive junction 37, constitute the further transfer gate 21.

[0068] The read-out junction 38 and the read-out gate 40, together with the photosensitive junction 37 constitute the transfer gate 26.

[0069] It is important to note that the retention of the signal at the node 28 is possible due to the presence of a capacitance connected to this node, which may be intrinsic due to the parasitics of the devices connected to this node or added to the architecture of the pixel 20.

[0070] With reference to FIG. 4 an embodiment variant of the pixel 20, shown in the representation of greater structural detail similar to that of FIG. 3 comprises, in addition to the elements of the embodiment of FIG. 3: a signal maintenance junction, 41 , which acts as a memory; a first transfer gate, 40a adapted to transfer the charge from the photosensitive junction 37 to the signal maintenance junction 41 ; a second transfer gate, 40b, adapted to transfer the charge from the signal maintenance junction 41 to the read-out junction 38.

[0071] In the purely exemplary embodiment of FIG. 1 , the photodetector 10 comprises a single pixel 20, a single discharge control circuit 50, and a single charge drainage component 60.

[0072] With reference to FIGS. 5 and 6, in accordance with one embodiment, a photodetector 10 according to the invention comprises a plurality of pixels 20, which have respective discharge conditioning terminals 22 connected to each other and to a single discharge control circuit 50, in turn connected to the drainage element 60.

[0073] With reference to FIGS. 7 to 9, in accordance with one embodiment, a photodetector 10 according to the invention comprises a plurality of pixels 20 and a plurality of discharge control circuits 50 (together 70), each associated with a single drainage element 60.

[0074] With reference to FIGS. 10 to 14, in accordance with one embodiment, a photodetector 10 according to the invention comprises a plurality of pixels 20 and a plurality of discharge control circuits 50 associated with a single charge drainage component 60.

[0075] With reference to FIGS. 5 to 14, an electro-optical sensor, 90, according to the present invention is formed by one or more photodetectors 10 according to one of the embodiments described above, wherein the relative pixels 20 form an array, 80. For simplicity, the figures do not represent the typical read-out and selection electronics of six CMOS sensors in matrix or linear formats.

[0076] In the figures, a group consisting of at least one pixel 20 and a discharge conditioning circuit 50 is indicated with 70.

[0077] According to an embodiment of the invention, of the two elements constituting the active charge transfer element, 62, 63, the current mirror element on the discharge circuit, 62, mirroring the current on the discharge conditioning circuit 50, may be included in the discharge conditioning circuit 50, instead of in the drainage element 60, and then be repeated for the same number of times as the discharge conditioning circuit 50 is repeated, so as to mirror the current individually for each of them, while the other mirror element, 63, is comprised in the charge drainage component 60. [0078] According to further embodiments of the present invention, the discharge control circuit 60 can be implemented with other solutions known to the state of the art such as, by way of example, a programmable digital analogue converter.

[0079] The configurations presented are only examples of possible architectures while many other aggregations and combinations are possible.

[0080] With reference to FIG. 1 , the pixel 20 is of the type suitable for detecting light with wavelengths belonging to the visible and near infrared spectrum, but embodiments in which the photosensitive material is such as to allow the detection of light even in other ranges such as ultraviolet or infrared are not excluded. In addition, the light intensity detectable with a single acquisition can reach, with the appropriate configuration, more than 6 decades.

[0081] The photosensitive receiving means 23 is a reverse polarized diode, which is advantageously made with an isolated N-type junction on P-type substrate, however embodiment variants of such a junction can be chosen from those known in the state of the art.

[0082] The discharge conditioning circuit 50, thanks to the connection with the further transfer gate 21 , makes it possible to vary the condition of the photosensitive receiving means 23 depending on the state of the signal applied to the discharge conditioning terminal 22, to the transfer terminal 25 and to the reset terminal 27. The photosensitive receiving means 23 can therefore be reset, when the terminals 25 and 27 allow a connection of the transfer gate 26 and the element useful for providing a reference and polarization voltage 30, respectively, and the photosensitive receiving means 23 subsequently enters an integration phase when the signal 25 decouples the read-out node 28 from the photosensitive receiving means 23. During the integration phase, the signal generated by the discharge conditioning circuit 50 and transmitted via the discharge conditioning terminal 22 to the further transfer gate 21 allows a controlled discharge of the photosensitive receiving means 23 according to a variable and programmable curve.

[0083] In the architecture shown in FIG. 3, the resetting mechanism of the photosensitive receiving means 23 is effected by a charge transfer to the read-out junction 38, which, if correctly performed, completely empties the photosensitive receiving means 23 into the reset condition. In the architecture shown in FIG. 4, the transfer takes place first at the signal holding junction 41 and then at the read-out junction 38 to allow the signal to be held in an isolated and light insensitive junction and enable the global shutter function that allows the data of all the pixels of the sensor to be frozen at the same time. [0084] FIGS. 15 and 16 show two possible signal sequences for the generation of the discharge curve by the discharge conditioning circuit 50 in which CLK indicates the timing signal on the basis of which all other signals are generated, VREF indicates the reference and external polarization signal, 55, BOOST indicates the signal transmitted to the boosting terminal, 54, SYNC indicates the synchronization signal 56 transmitted to the switch 53, HD indicates the generated discharge curve, i.e. the control signal transmitted from the discharge conditioning circuit 50 to the discharge conditioning terminal 22.

[0085] FIG. 17 finally shows some examples of discharge curves that can be generated by the discharge conditioning circuit 50 and applied to the discharge conditioning terminal 22. Reference (a) shows a conventional control signal wherein at the beginning of the integration time a pulse is provided. Reference (b) shows a progressively decreasing control signal. Reference (c) shows a progressively decreasing control signal of a linear type in sections. Reference (d) shows a progressively decreasing control signal mixed in steps and linear in sections. Reference (e) shows a progressively decreasing control signal in a linear manner. Reference (f) shows a programmed pulse type control signal.

[0086] It is also clear that, although the present invention has been described with reference to some specific examples, a person skilled in the art will certainly be able to make many other equivalent forms of photodetector devices 10, having the features expressed in the claims and therefore all falling within the scope of protection defined thereby.

Claims

1. Method for controlling the pixel discharge in electro-optical sensors wherein at least one pixel (20) comprises: at least one photosensitive receiving means (23), a read-out node (28) to which a read-out architecture (27, 30, 31 , 33, 34, 35) is connected, comprising an element (30) useful for providing the pixel (20) with a reference and polarization voltage, said element (30) being controlled by a reset terminal (27), said element (30) being arranged in a configuration decoupled from said photosensitive receiving means (23); a transfer gate (26) controlled by a transfer terminal (25), said transfer gate (26) being interposed between said at least one photosensitive receiving means (23) and the read-out node (28) so that the read-out node (28) is decoupled from the photosensitive receiving means (23); a further transfer gate (21) for transferring the charge from the coupled photosensitive receiving means (23) directly to the at least one photosensitive receiving means (23), said further transfer gate (21) being controlled by a discharge conditioning terminal (22); said method being characterized in that a control signal of variable type is applied to said discharge conditioning terminal (22).

2. Method for controlling the pixel discharge in electro-optical sensors according to claim 1 , wherein said read-out architecture (27, 30, 31 , 33, 34, 35) comprises: a read-out terminal (35), a switch (33) for connecting the rest of the pixel to the read-out terminal (35), said switch (33) being controlled via a selection terminal (34); an element useful for providing a reference and polarization voltage (30) to the pixel (20) controlled by a reset terminal (27); an active amplification and decoupling element of the signal (31) arranged between the read-out node (28) and the read-out terminal (35).

3. Method for controlling the pixel discharge in electro-optical sensors according to claim 1 or 2 wherein said control signal applied to said discharge conditioning terminal (22) is of a programmable type.

4. Method for controlling the pixel discharge in electro-optical sensors according to one of the preceding claims wherein said control signal is initially brought to a reference value, and then discharged progressively and in a controlled manner.

5. Photodetector (10) comprising at least one pixel (20) comprising: at least one photosensitive receiving means (23), a read-out node (28) to which a read-out architecture (27, 30, 31 , 33, 34, 35) is connected, comprising an element (30) useful for providing the pixel (20) with a reference and polarization voltage, said element (30) being controlled by a reset terminal (27), said element (30) being arranged in a configuration decoupled from said photosensitive receiving means (23); a transfer gate (26) controlled by a transfer terminal (25), said transfer gate (26) being interposed between said at least one photosensitive receiving means (23) and the read-out node (28) so that the read-out node (28) is decoupled from the photosensitive receiving means (23); a further transfer gate (21) for transferring the charge from the coupled photosensitive element (23) directly to the at least one photosensitive receiving means (23), said further transfer gate (21) being controlled by a discharge conditioning terminal (22); said photodetector (10) being characterized in that it comprises at least one discharge control circuit (50) suitable for bringing a variable voltage to said discharge conditioning terminal (22) so as to be able to dynamically condition the amount of charge dissipated by the incident light.

6. Photodetector (10) according to the preceding claim, characterized in that said readout architecture (27, 30, 31 , 33, 34, 35) comprises: a read-out terminal (35), a switch (33) for connecting the rest of the pixel to the read-out terminal (35), said switch (33) being controlled via a selection terminal (34); an element useful for providing a reference and polarization voltage (30) to the pixel (20) controlled by a reset terminal (27); an active amplification and decoupling element of the signal (31) arranged between the read-out node (28) and the read-out terminal (35).

7. Photodetector (10) according to the preceding claim, characterized in that it comprises at least one charge drainage element (60) associated with said discharge conditioning circuit (50).

8. Photodetector (10) according to the preceding claim, characterized in that said discharge control circuit (50) comprises: a charging element (51) which transmits a charging signal, a capacitance (52) capable of maintaining the charging signal transmitted by the charging element (51), and a switch (53) for selectively connecting the charge drainage component (60).

9. Photodetector (10) according to claim 5 or 6, characterized in that said charge drainage element (60) comprises: an active charge transfer element (62, 63) consisting of two elements in a current mirror configuration; a current generating element (65).

10. Photodetector (10) according to claim 4 or subsequent ones, characterized in that it comprises a plurality of pixels (20) associated with a single discharge conditioning circuit (50).

11. Photodetector (10) according to claim 4 or subsequent ones, characterized in that it comprises a plurality of discharge conditioning circuits (50) associated with a single charge drainage element (60).

12. Electro-optical sensor (90) characterized in that it comprises a plurality of photodetector devices (10) according to one of the preceding claims whose pixels (20) are arranged in such a way as to constitute an array of elements (80), driven and read by means of appropriate control and reading electronics.