Detector array with disturbance detection

Abstract

The invention relates to a detector array organised in rows and columns of pixels, the detector being sensitive to a variable physical effect, the detector comprising row conductors (Phi_ligne), column conductors (Col) connected to each of the pixels (P) and a readout circuit connected to each column conductor (Col), the pixels comprising a component (D) sensitive to the physical effect and a first switch (T1) connected between the sensitive component (D) and one of the column conductors (Col), the first switch (T1) being controlled by one of the row conductors (Phi_ligne), in at least one of the columns of pixels, the pixels of this column comprise a second switch (T2) that is controlled by one of the row conductors (Phi_ligne), connected via one of its terminals (T2a) to one of the column conductors (Col41) not connected to the first switches (T1), and coupled without connection to the sensitive component (D).

Description

DESCRIPTION Title of the invention: Matrix detector for disturbance detection

[0001] The invention relates to a matrix detector. The invention is particularly useful for creating a photosensitive detector used to generate visible images. The invention is not limited to the creation of this type of detector. It can also be used to create a detector capable of generating pressure or temperature maps, or two-dimensional representations of chemical or electrical potentials. These maps or representations form images of physical quantities.

[0002] The invention applies in particular to the development of passive or active matrix detectors used, for example, for detection purposes in devices that produce visible images. The physical phenomenon is then electromagnetic radiation, for example, in the visible wavelength band. The device can be adapted to detect ionizing radiation, for example, X-rays or gamma rays, either directly or through a scintillator.

[0003] In a matrix detector, a pixel represents the detector's elementary sensing element. Each pixel converts a physical phenomenon to which it is subjected into an electrical signal. The electrical signals from the different pixels are collected during a readout phase of the matrix and then digitized so that they can be processed and stored to form an image. The pixels comprise a component sensitive to the physical phenomenon and deliver, for example, a current of electrical charge. The physical phenomenon can be electromagnetic radiation carrying a flux of photons, and subsequently, the invention will be explained using this type of radiation, with the charge current being a function of the photon flux received by the sensitive component. Generalization to any type of matrix detector will be straightforward.

[0004] A matrix image detector comprises row conductors, each connecting the pixels of the same row, and column conductors, each connecting the pixels of the same column. The row conductors are connected to driver modules, usually located on one edge of the matrix, which control the individual pixels. The column conductors are connected to conversion circuits, also usually located on one edge of the matrix. The edge of the matrix can be called the "column foot." Conversion circuits allow the pixel matrix to be read. It is understood that the terms "rows" and "columns" are purely conventional and can be reversed.

[0005] Each pixel typically includes a sensitive component, such as a photodetector, which can be a photodiode, a photoresistor, or a phototransistor. Large photosensitive arrays exist, containing millions of pixels arranged in rows and columns. Each pixel also includes an electronic circuit with at least one switch. This electronic circuit may also include other switches, capacitors, and resistors, downstream of which the switch is located. The combination of the sensitive component and the electronic circuit generates and collects electrical signals. The electronic circuit may also reset the signal collected in each pixel after it has been read.The switch's role is to transfer or copy, within a column conductor, the signals collected by the electronic circuit based on the information received from the sensing component. This transfer occurs when the switch receives the instruction from a line conductor. The switch is connected, at one of its terminals, to the sensing component, either directly or indirectly, and at the other terminal, to the conversion circuit via one of the column conductors. This other terminal of the switch corresponds to the pixel output.

[0006] Matrix photosensitive detectors are inherently very sensitive to disturbances that may originate from the environment in which they are used. These disturbances can be mechanical (shocks, vibrations), electromagnetic (electric and / or magnetic fields from, for example, electric motors or radio frequencies emitted by radiating devices) or thermal (temperature variations inducing a change in the leakage currents of the photosensitive components).

[0007] These disturbances can alter the signal collected by the pixels and thus create defects in the images. These defects are generally correlated structures along lines due to a principle of pixel matrix reading when, for example, it is performed line by line. The defects can also have various appearances depending on the type of disturbance and the principle of signal detection and reading.

[0008] These disturbances are particularly difficult to eliminate, especially when the detector is used autonomously or when correction must be done in real time, for example in fluoroscopy mode in the case of medical imagers.

[0009] It is possible to assess a disturbance during the reading of a line or a portion of a line and to subtract the signal corresponding to an estimate of this disturbance. This assessment can use information acquired at a time other than that of a complete image capture cycle. The assessment can result from a calculation or a measurement. All methods that involve temporally separating the reading of the useful signal from the reading of a signal used to estimate the disturbance lead to the same difficulty related to the impossibility of accurately predicting the exact value of the disturbance from a prior or subsequent measurement.

[0010] It is also possible to use a double readout, with and without a useful signal, which has the major drawback of increasing the image acquisition time. This double readout is performed during the image cycle and is known as Correlated Double Sampling, or by its acronym CDS. It involves reading the information contained in the pixel array twice. The first readout is used to read the useful signal, and the second is used to read the disturbances without a useful signal. Subtracting the two readouts corrects the offset of the useful signal. This correction eliminates the offset drift generated by various external disturbances. However, this process reduces the overall readout speed of the detector by requiring two readouts of the array. This process also tends to increase the detector's power consumption.Furthermore, correlated double sampling cannot account for offset variations that may occur between the two readings.

[0011] Other predictive assessment methods based on prior knowledge of the disturbance allow for the estimation of a disturbance using a frequency-based approach, but do not allow for an exact determination of the actual level of disturbance.

[0012] Another method for correcting disturbances involves spatially separating the reading of the useful signal from the reading of the signal used to estimate the disturbance. This approach is feasible when the local variation of the disturbance is assumed to be small or negligible. It is then possible to simultaneously read the disturbed signal and a signal used to estimate the disturbance in two areas that are as close as possible. This method requires pixels dedicated to the disturbance estimation, often called masked pixels, which do not receive a useful signal, and it is not very precise in the local estimation of disturbances.

[0013] The common characteristic of all the methods mentioned above is to use information that is temporally or spatially offset from the desired information: either this information corresponds to the value of the disturbance at a different time, assuming that the variation is sufficiently small or well modeled, or it corresponds to the value of the disturbance at a different location but requires a double reading and therefore a lengthening of the time required to obtain the image.

[0014] The invention aims to overcome all or part of the problems mentioned above by providing a matrix detector that allows for the estimation of disturbances synchronously and without delaying image reading. More specifically, the invention relates to a matrix detector comprising at least one pixel configured to simultaneously acquire a useful signal and a disturbance signal. Thus, the disturbance can be measured without spatial offset from the location of useful signal acquisition. Furthermore, the disturbance is read synchronously with the reading of the useful signal.

[0015] To this end, the invention relates to a matrix detector organized into rows and columns of pixels, the detector being sensitive to a variable physical phenomenon, the detector comprising row conductors, column conductors connected to each pixel and a readout circuit connected to each column conductor, the pixels comprising: - a component sensitive to the physical phenomenon, configured to acquire information whose value varies according to the variation of the physical phenomenon, - a first switch connected between the sensing component and one of the column conductors, and configured to transfer the information acquired by the sensing component to the column conductor, the first switch being ordered by one of the line operators, - in at least one of the pixel columns, the pixels of that column include a second switch controlled by one of the line conductors, connected at one of its terminals to one of the column conductors not connected to the first switches and coupled without connection to the sensitive component.

[0016] Several columns of pixels, whose pixels include the second switch connected to one of the column conductors, can be distributed at regular intervals in the pixel matrix.

[0017] The column(s) of pixels whose pixels include the second switch connected to one of the column conductors can be arranged at the edge of the matrix.

[0018] The matrix detector can include columns of pixels that do not have a second switch.

[0019] Alternatively, all pixels of the detector can understand the second switch.

[0020] The first and second electronic switches are advantageously of the same nature.

[0021] The first and second electronic switches are advantageously identical.

[0022] The coupling of the second switch to the sensitive component is, for example, primarily a capacitive coupling.

[0023] Advantageously, for the detector as a whole: a second column conductor connected to a reference voltage is present between two consecutive first column conductors.

[0024] In each pixel of the detector, the sensitive component is connected between the first switch and the reference voltage. One of the second column conductors is advantageously used to connect the sensitive component to the reference voltage.

[0025] The invention will be better understood and other advantages will become apparent upon reading the detailed description of an embodiment given by way of example, a description illustrated by the accompanying drawing in which:

[0026] Figure 1 schematically represents an example of a matrix detector according to the invention;

[0027] Figure 2 represents a variant of the detector in Figure 1;

[0028] Figure 3 represents another variant of the detector in Figure 1.

[0029] For the sake of clarity, the same elements will carry the same markers in the different figures.

[0030] Figure 1 schematically represents a matrix detector with a two-row matrix (I1 and I2) and four columns (c1, c2, c3, and c4) for simplicity. Eight pixels are formed, each at the intersection of a row and a column. It is understood that real matrices are generally much larger and have a large number of rows and columns.

[0031] Each pixel comprises a sensitive component, represented here by a photodiode D, and an electronic processing circuit including an electronic switch, in the example shown, a transistor T1. Alternatives to a transistor can form the electronic switch. A diode, in particular, can be used. Furthermore, this type of pixel can include other components, notably other electronic switches.

[0032] A matrix detector according to the invention, based on photodiodes or phototransistors sensitive to light in the visible band, is generally combined, for X-ray or gamma radiography, with a scintillator, for example made of cesium iodide (Csl), enabling the conversion of X-ray or gamma photons into visible photons to which the photodiodes or phototransistors are sensitive. The combination of photodetectors and a scintillator ensures an indirect conversion of X-ray or gamma photons.

[0033] Alternatively, the invention can be implemented in a so-called direct conversion detector in which X-ray or gamma photons are directly converted into electrical charges in a conversion layer associated with a matrix. Each pixel of this matrix may include an associated storage capacitor, and, as in indirect conversion detectors, an electronic switch. The storage capacitor is itself associated with an electrode in contact with the conversion layer to locally receive the electrical charges resulting from the conversion of the received X-ray photons.

[0034] To date, several materials have been successfully tested for use in conversion coatings. These include cadmium telluride (CdTe), amorphous selenium (a:Se), thallium bromide (TIBr), perovskites, and lead oxides (PbO).

[0035] In a direct conversion detector, the storage capacity is considered as the component sensitive to the variable physical phenomenon that a matrix detector according to the invention can image.

[0036] Any other type of sensitive component can be implemented within the framework of the invention, such as a photoconductor.

[0037] In general, it is known to fabricate pixel arrays comprising transistors using complementary crystalline silicon semiconductors, known in English-language literature by their abbreviation CMOS for "Complementary Metal Oxide Semiconductor." The invention is not limited to this type of transistor; it can, for example, be implemented for arrays comprising thin-film field-effect transistors, known in English-language literature as TFTs for "Thin-film transistors." TFT transistors can be based on metal oxides, such as amorphous or crystalline indium, gallium, and zinc oxide transistors, known by their English abbreviation IGZO. Other families of TFT transistors can be used, such as organic TFTs, amorphous silicon TFTs, and polycrystalline silicon TFTs.

[0038] The pixels in a single row are connected to a line conductor, respectively Phijignel and Phi_ligne2, which carries a signal to control each row of pixels. In the case of pixels with additional transistors, these can be controlled by other line conductors. For example, each pixel might include another electronic switch to reset photodiode D.

[0039] The pixels in the same column are connected to a column conductor, respectively Col1, Col2, Col3, and Col4, allowing the charges accumulated by the photodiode D to be read, line by line. Each column conductor Col1 is connected to a readout circuit L configured to read the signal level accumulated by the sensitive components connected to it when the transistor T1, connected The signal is conducting at the sensitive component in question. Each reading circuit can be configured to read an electrical signal present in the column to which it is connected. This signal is representative of the level of the physical phenomenon detected by the sensitive component. The signal can be an electrical charge, a voltage, a current, etc.

[0040] In the example shown, where the sensitive component is a photodiode and the electronic circuit consists only of transistor T1, the signal representing the illumination received by the photodiode is formed from the electrical charges read by the corresponding readout circuit. The electronic circuit may also include a follower transistor configured to convert the electrical charges accumulated by the photodiode into a voltage. This additional transistor is placed between the photodiode and transistor T1. The follower transistor is often paired with a photodiode reset transistor. This type of pixel is often called a 3T pixel, as opposed to the 1T pixel shown in Figure 1.

[0041] During an image acquisition phase, the illumination received by photodiode D causes the potential of its cathode to decrease. This image acquisition phase is followed by a readout phase during which the potential of photodiode D is read. To do this, transistor T1, which acts as a switch, is turned on by the control signal carried on conductor Phijigne and applied to its gate. The readout of photodiode D drains the charges accumulated on its cathode, thus resetting the pixel before the next image acquisition phase.

[0042] According to the invention, in at least one of the pixel columns, for example column c4, the pixels of this column comprise a switch T2 coupled without connection to the sensitive component. The switch T2 is connected at one of its terminals T2a to a column conductor Col41 not connected to the first switches T1.

[0043] We saw earlier that the column conductors Coll, Col2, Col3, and Col4 connect the corresponding electronic switches T1 to readout circuits to read the useful signals from the corresponding pixels. Similarly, the column conductor Col41 connects the electronic switches T2 of pixel column c4 to a readout circuit L, allowing the reading of the disturbance collected by each of the electronic switches T2. Disturbance reading is done line by line as for reading the useful signals accumulated by the D photodiodes.

[0044] The read circuits L are advantageously all identical. The read circuits can be arranged on the same substrate as that carrying the pixels or on a separate substrate arranged on an edge of the matrix.

[0045] In the example shown in Figure 1, the electronic switches T2 of a single column of pixels, column c4, are connected to the column conductor Col41 to enable readout. Column c4 can be located on an edge of the matrix. This allows the spacing between the column conductors to remain constant, facilitating connection to the readout circuits L. It is, of course, possible to arrange columns of pixels within the matrix where the electronic switches T2 are connected to readout circuits L via column conductors. These pixel columns, where disturbances are readout, can be arranged at regular intervals, for example, one column in ten. The column conductors are connected either to electronic switches T1 or electronic switches T2. It is preferable not to change the spacing between pixels, as this results in a non-constant spacing of the column conductors.The connection to the L reading circuits is adapted according to this step.

[0046] The electronic switches, and in particular the transistors T1 and T2, are advantageously of the same type, either a diode or a transistor, and even more advantageously identical except for their connections. Thus, the two electronic switches T1 and T2 have the same sensitivity to disturbances. Furthermore, their identical nature simplifies the design and implementation of the detector.

[0047] The proximity of photodiode D and transistor T2 results in a primarily capacitive coupling between them. It is also possible to extend the terminal of transistor T2, which is not connected to column conductor Col41, by means of a trace running along photodiode D, thereby adding inductive coupling to the capacitive coupling.

[0048] Switch T2 is controlled by the signal carried by one of the line conductors. Within the same pixel of column c4, it is perfectly possible to control switches T1 and T2 by the same line conductor Phijigne. Alternatively, as illustrated in the example shown in Figure 1, Switch T1 is connected to the line conductor of the same rank, and switch T2 is connected to the line conductor of the next rank. For example, for the pixel labeled P(1,4) located in row 1 and column 4, the gate of transistor T1 is connected to the line conductor Phi_ligne1, and the gate of transistor T2 is connected to the line conductor Phi_ligne2. This simplifies the placement and connection of the various components for each pixel.

[0049] During the read phase, for column c4, transistor T1 and transistor T2 are read simultaneously. As mentioned above, for the connection of the gates of transistors T1 and T2 to their respective line conductors, the transistors T1 and T2 read simultaneously can belong to the same pixel or to two adjacent pixels. In both cases, the simultaneous readout allows obtaining so-called useful information transferred by transistor T1 and information representing disturbances in the immediate vicinity of the photodiode. At the level of the readout circuits, or downstream of them, it is possible to subtract the information representing disturbances from the useful information, in order to obtain useful information free of these disturbances. The invention makes it possible to obtain information representing disturbances of one pixel at a time at the same instant as the useful information and in a spatially close vicinity.

[0050] To achieve perfect simultaneity between useful information and information representing disturbances, it is possible to subtract the information from two adjacent pixels on two neighboring lines. In the example in Figure 1, the information from transistor T2 on line 11 is subtracted from the information from transistor T1 on line I2. Indeed, the same line conductor, Phi_line2, controls both transistor T1 on line I2 and transistor T2 on line 11, which ensures perfect simultaneity between the readings of these two transistors.

[0051] The subtraction operation is, of course, only given as an example. It is possible to implement a more complex operation to correct the useful signal from transistor T1 using the signal from an adjacent transistor T2.

[0052] In the example shown in Figure 1, all pixels include transistor T2. This allows all pixels in the detector to have the same components, namely the photodiode D and the two transistors T1 and T2. As As mentioned above, the pixels can also include other components that are then found in all pixels, regardless of whether the T2 transistors are connected to a column conductor or not. This improves the homogeneity of the different detector pixels, particularly in terms of interference.

[0053] For pixel columns containing T2 transistors not connected to a column conductor, these T2 transistors can be left unconnected. Alternatively, to prevent their potential from floating, it is advantageous to connect them, for example, to a reference voltage as shown in Figure 1. Furthermore, in each pixel, the anode of photodiode D is connected to the reference voltage Vref. It is advantageous to connect the T2 transistors to this same voltage, which is already available.

[0054] Figure 2 shows an alternative variant in which only the pixels in column c4 contain T2 transistors. The other pixel columns do not contain T2 transistors. This simplifies the detector design. As mentioned above, this simplification may introduce differences in behavior between pixels equipped with T2 transistors and those without.

[0055] In the detectors shown in Figures 1 and 2, the individual pixels are connected to the reference voltage Vref by means of a special column conductor called Cref. Alternatively, the reference voltage Vref can be supplied to the individual pixels by means of a line conductor, or even a conductive plane placed beneath the pixel array. However, using column conductors to supply the reference voltage limits the coupling that can occur between two consecutive column conductors Col. Therefore, between two column conductors Col connected to readout circuits, it is advantageous to systematically include, for the entire pixel array, a column conductor Cref connected to the reference voltage Vref.

[0056] It is of course possible to have a column conductor Cref not connected to pixel components and yet allowing the limitation of coupling between two consecutive column conductors Col.

[0057] The various figures represent examples of matrix detector diagrams. In these diagrams, the relative position of the different components and the conductors connecting them is, however, representative of actual detector implementations.

[0058] In the variants shown in Figures 1 and 2, only one column of pixels, located on one edge of the matrix, contains a transistor T2 connected to a readout circuit through column conductor Col41. It is then easy to create pixels where the column conductors Col, which read the photodiode D, are all located on the same side of the pixels: to the left of the pixel in the example shown. Column Col41, connected to a readout circuit L and allowing the disturbance to be read through transistor T2, can then be located on the other side of column c4 of pixels, on the right as shown. It is then possible to place column conductor Cref of column c4 between column conductor Col4 and column conductor Col41 to limit the coupling between the two column conductors Col4 and Col41.

[0059] Figure 3 illustrates a detector variant in which columns of pixels equipped with T2 transistors connected to readout circuits are not arranged at the edge of the matrix. By arranging all the pixel columns so that the column conductors connected to their respective T1 transistors are all located on the same side of the pixels, the column conductors (Col) would necessarily be placed in close proximity to one another without any intervening column conductors (Cref). Figure 3 illustrates the advantageous presence of a Cref conductor separating two consecutive Col conductors, even around a column of pixels equipped with T2 transistors connected to a readout circuit, this column of pixels being arranged between two other columns of standard pixels, i.e., those not equipped with T2 transistors connected to readout circuits.

[0060] Two arrangements, illustrated in Figure 3, allow a column of pixels equipped with T2 transistors connected to a readout circuit to be placed at the heart of the pixel matrix.

[0061] One approach involves arranging the column conductors in a mirrored fashion in two adjacent columns of pixels, specifically columns c2 and c3. More precisely, columns c1 and c2 do not have T2 transistors. connected to readout circuits, the column conductors Coll and Col2 are positioned to the left of their respective pixel columns. For column c3, the column conductor Col31, which reads the T2 transistors, is positioned to the left of the c3 pixel column, and the column conductor Col3 is positioned to the right of the c3 pixel column. Higher-rank pixel columns can retain the arrangement of the c3 pixel column, with its column conductor connected to transistors T2 positioned to the left and its column conductor connected to transistors T1 positioned to the right. It is therefore possible to place a column conductor Cref between all the column conductors Col, whether they are column conductors connected to transistors T1 or column conductors connected to transistors T2.

[0062] However, this mirrored arrangement only allows for a single column of pixels at the core of the matrix, where the T2 transistors are connected to read circuits. To add more columns of pixels where the T2 transistors are read, it is possible to place additional Cref column conductors between two columns of pixels within the matrix, without connecting them to any neighboring pixels. This arrangement is illustrated in Figure 4 between the pixel columns c3 and c4.This arrangement allows column c4 to return to an arrangement identical to that of pixel columns c1 and c2, namely with the column conductor Col4 positioned on the left. This reversal of pixel column c4 allows the pattern of the succession of columns c1 to c4 to be reproduced periodically, and thus periodically inserts columns of pixels into the matrix whose T2 transistors are connected to their respective readout circuits, while maintaining a column conductor Cref systematically positioned between all pairs of consecutive column conductors Col. The periodicity of the pixel columns whose T2 transistors are connected to readout circuits must be determined empirically based on the spatial evolution of the perturbations.

Claims

DEMANDS 1. Matrix detector organized in rows and columns of pixels, the detector being sensitive to a variable physical phenomenon, the detector comprising line conductors (Phijigne), first column conductors (Col) connected to each pixel and a readout circuit (L) connected to each first column conductor (Col), the line conductors (Phijigne) and the first column conductors (Col) being connected to each pixel, each pixel comprising: - a sensitive component (D) to the physical phenomenon, configured to acquire information whose value varies according to the variation of the physical phenomenon, - a first switch (T1) connected between the sensitive component (D) and a first of the column conductors (Col), and configured to transfer the information acquired by the sensitive component (D) to the first column conductor (Col), the first switch (T1) being controlled by one of the line conductors (Phijigne), in at least one of the pixel columns, each of the pixels in this column includes a second switch (T2) controlled by one of the line conductors (Phijigne), connected at one of its terminals (T2a) to a first of the column conductors (Col31; Col41) not connected to the first switches (T1) and coupled without connection to the sensitive component (D).

2. Matrix detector according to claim 1, in which several columns of pixels whose pixels comprise the second switch (T2) connected to one of the first column conductors (Col31; Col41) are distributed at regular intervals in the pixel matrix.

3. Matrix detector according to claim 2, in which the pixel column(s) whose pixels comprise the second switch (T2) connected to one of the first column conductors (Col31; Col41) are arranged at the edge of the matrix.

4. Matrix detector according to any one of the preceding claims, wherein the matrix detector comprises columns of pixels (c1, c2, c3, c4) not having a second switch (T2).

5. Matrix detector according to any one of claims 1 to 3, wherein all pixels of the detector include the second switch (T2).

6. Matrix detector according to any one of the preceding claims, wherein the first (T1) and the second (T2) electronic switches are of the same nature.

7. Matrix detector according to any one of the preceding claims, wherein the first (T1) and second (T2) electronic switches are identical.

8. Matrix detector according to any one of the preceding claims, wherein the coupling of the second switch (T2) to the sensitive component (D) is mainly a capacitive coupling.

9. Matrix detector according to any one of the preceding claims, wherein for the whole of the detector: a second column conductor (Cref) connected to a reference voltage (Vref) is present between two consecutive first column conductors (Col).

10. Matrix detector according to the preceding claim, in each of the pixels, the sensitive component (D) is connected between the first switch (T1) and the reference voltage (Vref) carried by one of the second column conductors (Cref).

Images