KR102016570B1 - Touch sensing system and noise reduction method thereof - Google Patents

Abstract

The present invention relates to a touch sensing system and a method of removing noise thereof, the method comprising: converting signals received from touch sensors into digital data to generate touch raw data; Calculating local statistical parameters including an average value and a variance value of Nth (N is a positive integer) -2 to Nth data by mapping the touch furnace data to a preset window; Calculating filter coefficients of a local Gaussian filter based on the local statistical parameters; And after correcting the N-th data using the local Gaussian filter, shifting the window to correct other data.

Description

TOUCH SENSING SYSTEM AND NOISE REDUCTION METHOD THEREOF}

The present invention relates to a touch sensing system and a method for removing the noise.

A user interface (UI) enables communication between a person (user) and various electric and electronic devices, so that the user can easily control the device as desired. Representative examples of the user interface include a keypad, a keyboard, a mouse, an On Screen Display (OSD), a remote controller having infrared communication or radio frequency (RF) communication. User interface technology continues to evolve in the direction of increasing user sensitivity and ease of operation. In recent years, user interfaces have evolved into touch UIs, voice recognition UIs, 3D UIs, and the like.

Touch UI is being adopted to portable information devices, and it is being applied to home appliances. The capacitive touch sensing system has the advantage that the touch screen structure has higher durability and clarity than the conventional resistive method, and can be applied to various applications.

Analog signals received from the touch sensors of the touch screen include a sensor signal proportional to a driving signal applied to the touch sensor and noise as shown in FIG. 1. Noise is divided into offset noise and random noise. The offset noise is mainly generated due to a change in the driving signal applied to the touch sensor. The offset noise cannot be accurately predicted in the frequency, time, and magnitude directions, but can be roughly measured as the average value of the touch furnace data during one frame period. Random noise is interference noise generated due to coupling with the display panel and noise generated due to manufacturing process variation. Noise added to the sensor signal lowers the signal-to-noise ratio (SNR), reducing touch sensitivity. Therefore, noise must be removed to increase the touch sensitivity.

The touch sensing system supplies driving signals to the touch sensors and converts the analog signals obtained from the touch sensors into digital data using an analog-to-digital converter (hereinafter referred to as "ADC") to convert the touch raw data into touch data. touch raw data is generated and the touch row data is analyzed to calculate the presence or absence of touch input and the coordinates of the touch input position. Korean Patent Laid-Open Publication No. 10-2007-0109360 (Nov. 15, 2007) proposed a method of removing offset noise using an average value of touch low data using an average filter. The patent also claims that random noise can be removed by the average of the signals applied to the same touch sensor. However, the noise canceling method disclosed in this publication can remove some of the offset noise by subtracting the average value of the touch row data from the touch row data, but there is almost no random noise reduction effect.

The present invention provides a touch sensing system capable of removing noise and a method of removing the noise.

The touch sensing system of the present invention includes a touch sensing circuit for converting signals received from touch sensors into digital data to generate touch raw data; And calculating a local statistical parameter including an average value and a variance value of the Nth (N is a positive integer) -2 to the Nth data by mapping the touch-low data to a preset window, and using a local Gaussian filter based on the local statistical parameter. And a noise removing unit for calculating a filter coefficient of, correcting the N-th data using the local Gaussian filter, and then shifting the window to correct other data.

The method for removing noise of the touch sensing system may include converting signals received from touch sensors into digital data and generating touch raw data; Calculating local statistical parameters including an average value and a variance value of Nth (N is a positive integer) -2 to Nth data by mapping the touch furnace data to a preset window; Calculating filter coefficients of a local Gaussian filter based on the local statistical parameters; And after correcting the N-th data using the local Gaussian filter, shifting the window to correct other data.

The present invention calculates the filter coefficients of the local Gaussian filter based on the local statistics of the touch furnace data defined in the window of the preset size, calculates the filter coefficients of the local Gaussian filter, and then uses the local Gaussian filter. Remove noise from As a result, the present invention can effectively remove noise that is difficult to remove in the average filter of the prior art to improve the signal sensitivity to improve the signal-to-noise ratio (SNR) in the touch sensing system.

1 is a diagram illustrating a sensor signal and noise in a signal received from a touch sensor.
2 is a diagram illustrating a touch sensing system according to an embodiment of the present invention.
3 is an enlarged view of a portion of the touch screen illustrated in FIG. 2.
4 to 6 illustrate various combinations of a display panel and a touch screen.
7 is a flowchart illustrating a control procedure of a noise removing unit according to an exemplary embodiment of the present invention.
8 is a diagram illustrating an example of a window for defining a local statistical area.
9 illustrates an example of an operation of a local Gaussian filter according to an exemplary embodiment of the present invention.
FIG. 10 is a diagram showing an experimental result comparing a treatment result of a local Gaussian filter of the present invention and a mean filter of the related art. FIG.
FIG. 11A is an enlarged view of a portion of the touch area 'Touch' in FIG. 10.
FIG. 11B is an enlarged view of a portion of the non-touch area 'Non Touch' in FIG. 10.

The touch sensing system of the present invention may be implemented as a capacitive touch screen that senses a touch input through a plurality of capacitive sensors. The capacitive touch screen includes a plurality of touch sensors. Each of the touch sensors includes a capacitance when viewed in equivalent circuitry. The capacitance can be divided into self capacitance or mutual capacitance. In the following embodiments, a mutual capacitive touch screen is illustrated, but is not limited thereto. Since the touch sensing system of the present invention has a feature of removing noise of a signal received from touch sensors, the touch sensing system may be implemented by any conventional touch sensing system.

The display device of the present invention is a liquid crystal display (LCD), a field emission display (FED), a plasma display panel (PDP), an organic light emitting diode display (Organic Light Emitting Display) , OLED), and electrophoretic display devices (Electrophoresis, EPD). In the following embodiment, although the liquid crystal display device is described as an example of a flat panel display device, it should be noted that the display device of the present invention is not limited to the liquid crystal display device and can be implemented with all known display devices.

 Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. Like numbers refer to like elements throughout. In the following description, when it is determined that a detailed description of known functions or configurations related to the present invention may unnecessarily obscure the subject matter of the present invention, the detailed description thereof will be omitted.

2 to 6, the touch sensing system of the present invention includes a touch screen TSP on which touch sensors Cts are arranged, a touch screen driving circuit, and the like. The touch screen TSP is bonded on the upper polarizing plate POL1 of the display panel DIS as shown in FIG. 4, or is formed between the upper polarizing plate POL1 and the upper substrate GLS1 of the display panel DIS as shown in FIG. 5. Can be. In addition, the touch sensors Cts of the touch screen TSP may be embedded in the lower substrate in an in-cell type together with the pixel array in the display panel DIS as shown in FIG. 6. 4 to 6, "PIX" means a pixel electrode of a liquid crystal cell, "GLS2" means a lower substrate, and "POL2" means a lower polarizing plate, respectively.

The display panel DIS includes a liquid crystal layer formed between two substrates. The pixel array of the display panel DIS includes pixels formed in the pixel area defined by the data lines D1 to Dm, where m is a positive integer, and the gate lines G1 to Gn, where n is a positive integer. . Each of the pixels is connected to TFTs formed at intersections of the data lines D1 to Dm and the gate lines G1 to Gn, a pixel electrode to charge a data voltage, and a pixel electrode. Storage capacitors (Cst) for maintaining the voltage and the like.

Black matrices, color filters, and the like are formed on the upper substrate of the display panel DIS. The lower substrate of the display panel DIS may be implemented in a color filter on TFT (COT) structure. In this case, the color filter may be formed on the lower substrate of the display panel DIS. The common electrode supplied with the common voltage may be formed on the upper substrate or the lower substrate of the display panel DIS. A polarizing plate is attached to each of the upper and lower substrates of the display panel DIS, and an alignment layer for setting the pretilt angle of the liquid crystal is formed on an inner surface of the display panel DIS. A column spacer is formed between the upper substrate and the lower substrate of the display panel DIS to maintain a cell gap of the liquid crystal cell.

The backlight unit may be disposed under the rear surface of the display panel DIS. The backlight unit is implemented as an edge type or direct type backlight unit to emit light to the display panel DIS. The display panel DIS may be implemented in any known liquid crystal mode, such as twisted nematic (TN) mode, vertical alignment (VA) mode, in plane switching (IPS) mode, or fringe field switching (FFS) mode.

The display driving circuit includes the data driving circuit 12, the scan driving circuit 14, and the timing controller 20 to write the video data voltage of the input image to the pixels of the display panel DIS. The data driving circuit 12 converts the digital video data RGB input from the timing controller 20 into an analog positive / negative gamma compensation voltage and outputs a data voltage. The data voltage output from the data driver circuit 12 is supplied to the data lines D1 to Dm. The scan driving circuit 14 sequentially supplies gate pulses (or scan pulses) synchronized with the data voltages to the gate lines G1 to Gn to select a line of the display panel DIS to which the data voltages are written.

The timing controller 20 inputs timing signals such as a vertical synchronization signal Vsync, a horizontal synchronization signal Hsync, a data enable signal Data Enable, DE, and a main clock MCLK input from the host system 50. In response, the operation timings of the data driver circuit 12 and the scan driver circuit 14 are synchronized. The data timing control signal for controlling the data driving circuit 12 includes a source sampling clock (SSC), a polarity control signal (Polarity, POL), a source output enable signal (SOE), and the like. do. The scan timing control signal for controlling the scan driving circuit 14 includes a gate start pulse (GSP), a gate shift clock, a gate output enable signal (GOE), and the like. do.

The host system 50 may be implemented as any one of a television system, a set top box, a navigation system, a DVD player, a Blu-ray player, a personal computer (PC), a home theater system, and a phone system. The host system 50 includes a system on chip (SoC) having a built-in scaler to convert digital video data RGB of an input image into a format suitable for displaying on the display panel DIS. The host system 50 transmits timing signals Vsync, Hsync, DE, and MCLK together with the digital video data to the timing controller 20. In addition, the host system 50 executes an application program associated with the coordinate information XY of the touch input position input from the touch screen driving circuit.

The touch screen TSP includes Tx lines (Tx1 to Txj, j is a positive integer smaller than n), and Rx lines (Rx1 to Rxi and i are positive amounts smaller than m) intersecting the Tx lines (Tx1 to Txj). Integer), and i × j touch sensors Cts formed at intersections of the Tx lines Tx1 to Txj and the Rx lines Rx1 to Rxi. Each of the touch sensors Cts includes mutual capacitance.

The touch screen driving circuit includes a touch sensing circuit 30, a noise removing unit 36, an algorithm executing unit 40, and the like. The touch screen driving circuit receives a charge of the touch sensor by supplying a driving signal to the touch sensors and removes noise by using a local Gaussian filter described later in the received signal. The touch screen driving circuit compares a signal passing through the local Gaussian filter with a predetermined threshold, detects a touch input having a threshold value or more, and executes a preset touch coordinate algorithm to calculate coordinates of the touch input position.

The touch sensing circuit 30 may include a Tx driver 32, an Rx sensor 34, a timing generator 38, and the like. The touch sensing circuit 30 applies a driving signal to the touch sensors Cts through the Tx lines Tx1 to Txj using the Tx driver 32, and synchronizes the Rx lines Rx1 to Rxi with the driving signal. And the Rx sensing unit 34 senses the voltages of the touch sensors Cts and outputs touch raw data. The driving signal may be generated in various forms such as a pulse, a sinusoidal wave, and a triangle wave. The touch sensing circuit 30 may be integrated into one read-out integrated circuit (ROIC).

The Tx driver 32 selects a Tx channel to output a drive signal in response to the Tx setup signal from the timing generator 38, and applies a drive signal to the Tx lines Tx1 to Txj connected to the selected Tx channel. . The Tx lines Tx1 to Txj supply charges to the touch sensors Cts during the high potential period of the driving signal. The driving signal is connected to the Tx lines Tx1 to Txj such that the voltages of the touch sensors Cts are accumulated on the capacitor of the integrator built in the Rx sensing unit 34 through the Rx lines Rx1 to Rxi. N (N is a positive integer above) can be supplied continuously to each.

The Rx sensing unit 34 selects Rx lines to receive the voltage of the touch sensor in response to the Rx setup signal from the timing generator 38 and selects Rx lines of the touch sensor Cts through the selected Rx lines in synchronization with the driving signal. Receive and sample the output voltage. The Rx sensing unit 34 accumulates the sampled voltage in a capacitor of an integrator, converts the voltage of the capacitor into digital data by using an analog-to-digital converter (hereinafter, referred to as "ADC"). Output digital data as touch raw data.

The timing generator 38 controls the Tx channel and the Rx channel settings, and synchronizes the Tx driver 32 and the Rx sensing unit 34.

The noise removing unit 36 maps a window having a predetermined size to touch raw data and removes noise by applying a local Gaussian filter within the window. The noise removing unit 36 removes noise from the data with another touch while shifting the window. The noise removing unit 36 will be described in detail with reference to FIGS. 7 to 9.

The algorithm executing unit 40 executes a preset touch coordinate algorithm to compare the touch row data passing through the noise removing unit 36 with a preset threshold. Any known algorithm may be used as the touch coordinate algorithm. The touch coordinate algorithm detects data with a touch above a threshold. The touch row data having a threshold value or more is determined as touch data obtained from touch sensors in which a touch input is generated. The algorithm execution unit 40 executes a touch coordinate algorithm to assign an identification number to each of the touch data having a threshold value or more and calculate coordinates for each of the touch input positions. The algorithm execution unit 40 executes a labeling algorithm so as to distinguish each of the touch input areas having a threshold value or more, and adds an identification code to each of the touch input areas, and applies the identification code to coordinate information of the touch input areas. XY) together with the host system 50. The noise removing unit 36 and the algorithm executing unit 40 may be implemented in one micro controller unit (MCU).

7 is a flowchart showing a control procedure of the noise removing unit 36 according to the embodiment of the present invention. 8 is a diagram illustrating an example of a window for defining a local statistical area. 9 illustrates an example of an operation of a local Gaussian filter according to an exemplary embodiment of the present invention.

7 to 9, the noise removing unit 36 receives data by touch from the touch sensing circuit 30 and defines a local statistical region by mapping a predetermined window to the data. (S1 and S2 ) Window may be a three tap filter window as shown in FIGS. 8 and 9. The three tap filter window defines the data with three consecutive touches. The three touch furnace data include pre-correction N (N is a positive integer) data p1 to be filtered and neighboring post-correction N-1 and N-2 data q2 and q1. . When the left is shifted from left to right, the N-th data q1 is the previous data and the N-th data q2 is the preceding data. The N-th and N-th data q1 and q2 are data from which noise is removed by applying a local Gaussian filter before the N-th data p1. The window is not limited in size to FIGS. 8 and 9. For example, if i touch sensors are arranged on one line of the touch screen TSP as shown in FIG. 1, the window may be a M (M is a positive integer greater than 3 and less than i / 2) tap filter windows.

The noise removing unit 36 calculates parameters by calculating local statistics necessary for Gaussian calculation. The parameters include an average value μ _p1 and a variance value σ of the local region defined by the window (S3 and S4). The noise removing unit 36 then uses Equation 3 as a function of the parameters. Compute the coefficient H (i) of the local Gaussian filter (S5).

Here, T _N is a noise control parameter, k is a threshold set to a value greater than 0 and less than 1. k can be adjusted according to the strength of the local Gaussian filter. ND is defined as the sum of absolute values of the difference between the previous value R (t-1, i) and the current value R (t, i) of the N-th data p1 to be filtered.

Subsequently, the noise removing unit 36 substitutes the filter coefficients and the data p1, q1 and q2 calculated in step S5 into Equation 4 to calculate the N-th data P ₁ from which the noise is removed. The noise removing unit 36 corrects the N-th data P ₁ through steps S1 to S6, and then shifts the window to the right by 1 data as shown in FIG. 9 and transfers the corrected N-th data P ₁ to the previous data ( q1).

The operation of the noise removing unit 36 will be described with reference to the example of FIG. 9. In this example, the window was set to a three tab filter window. In this example, the N-th through N-th data Q2 ', Q1', and P1 'are 252, 246, and 222 in the (t-1) th frame period. In the t-th frame period, the N-th through N-th data q2, q1, and p1 are 250, 252, and 254. When data as shown in FIG. 9 is input, the noise removing unit 36 calculates local statistical parameters μ _p1 and σ as follows. In this example, k was set to 0.0234375.

The noise removing unit 36 substitutes parameters into Equation 3 to calculate a filter coefficient of the local Gaussian filter as follows.

The noise removing unit 36 substitutes the above filter coefficients into Equation 4 to correct the N-th data p1 as follows.

P ₁ = (1 × 254 + 1.84 × 252 + 1.09 × 250) / (1 + 1.84 + 1.08) = 252.6

FIG. 10 shows an experimental result comparing the results of processing of a local Gaussian filter applied to the noise removing unit 36 of the present invention and a mean filter used as a noise filter in the prior art. FIG. 11A is an enlarged view of a portion of the touch area 'Touch' in FIG. 10, and FIG. 11B is an enlarged view of a portion of the non-touch area 'Non Touch' in FIG. 10. As a result of comparing the mean square error of the data processed by the conventional Gaussian filter with the local Gaussian filter of the present invention, the local Gaussian filter of the present invention improves the signal-to-noise ratio (SNR) to the average filter. In comparison, the average noise error was reduced by about 3 times.

Those skilled in the art will appreciate that various changes and modifications can be made without departing from the technical spirit of the present invention. Therefore, the technical scope of the present invention should not be limited to the contents described in the detailed description of the specification but should be defined by the claims.

DIS: Display panel TSP: Touch screen
12: data driving circuit 14: scan driving circuit
20: timing controller 30: touch screen drive circuit
32: Tx drive unit 34: Rx sensing unit
36: noise removing unit 38: timing generator
40: algorithm execution unit

Claims (7)

A touch sensing circuit converting signals received from the touch sensors into digital data and generating touch raw data; And
Computes a local statistical parameter including an average value and a variance value of Nth (N is a positive integer) -2 to Nth data by mapping the touch raw data to a preset window, and calculates a local Gaussian filter based on the local statistical parameter. A noise removing unit for calculating filter coefficients, correcting N-th data using the local Gaussian filter, and then shifting the window to correct other data;
The noise removing unit,
And an average value (μ _p1 ) and a dispersion value (σ) of the N-th to N- _th data defined by the window are calculated by the following equation.

Where p ₁ is the N-th data before correction, q ₁ is the N-th data after correction, and q ₂ is the N-th data after correction. delete The method of claim 1,
The noise removing unit,
And a correction value (P ₁ ) of the N-th data by calculating a filter coefficient (H (i)) of the local Gaussian filter by the following equation.

Where T _N is a noise control parameter, k is a threshold set to a value greater than 0 and less than or equal to 1, and ND is a previous value (R (t-1, i)) and a current value (R) of the N-th data to be filtered. The sum of the absolute values of the difference of (t, i)).

Converting signals received from the touch sensors into digital data to generate touch raw data;
Calculating local statistical parameters including an average value and a variance value of Nth (N is a positive integer) -2 to Nth data by mapping the touch furnace data to a preset window;
Calculating filter coefficients of a local Gaussian filter based on the local statistical parameters; And
After correcting N-th data using the local Gaussian filter, shifting the window to correct other data;
Computing the local statistical parameter
And calculating an average value (μ _p1 ) and a dispersion value (σ) of the N-th to N- _th data defined by the window by the following equation.

Where p ₁ is the N-th data before correction, q ₁ is the N-th data after correction, and q ₂ is the N-th data after correction. delete The method of claim 4, wherein
Computing the filter coefficients of the local Gaussian filter,
The noise reduction method of the touch sensing system, characterized in that to calculate the filter coefficient (H (i)) of the local Gaussian filter.

Where T _N is a noise control parameter, k is a threshold set to a value greater than 0 and less than or equal to 1, and ND is a previous value (R (t-1, i)) and a current value (R) of the N-th data to be filtered. The sum of the absolute values of the difference of (t, i)). The method of claim 6,
After correcting the N-th data using the local Gaussian filter, correcting the other data by shifting the window,
Noise generation method of the touch sensing system, characterized in that for generating the correction value (P ₁ ) of the N-th data in the following equation.

Images