JP6858646B2 - Touch detection device and touch detection method - Google Patents

Description

The present invention relates to a touch detection device and a touch detection method capable of simultaneously detecting a plurality of touches.

A touch panel capable of specifying the position coordinates touched by the user is widely adopted for displays and mobile terminals as an excellent user interface means. Typical methods of such a touch panel include, for example, a resistance film method and a capacitance method. In particular, a projection-type capacitive touch panel can detect a plurality of touches at the same time, so that various user operations such as zooming in / out and rotating can be realized.

For example, the capacitance type touch panel described in Patent Document 1 has a plurality of row electrodes and column electrodes facing each other with an insulating layer interposed therebetween, and is a touch sensor provided at an intersection of the row electrodes and the column electrodes. Multi-touch is detected by measuring the maximum of the signal distribution based on the change in capacitance.

Japanese Unexamined Patent Publication No. 2015-125569

In recent years, as displays have become larger, touch panels used for displays have also become larger. On the other hand, it is difficult to increase the number of touch sensors used for the touch panel because the power consumption and the manufacturing cost increase. Therefore, as the touch panel becomes larger, the arrangement interval of the touch sensors becomes larger, and the touch resolution of the touch panel tends to decrease.

The technique described in Patent Document 1 has a problem that when the touch panel becomes large and the arrangement interval of the touch sensors becomes equal to or longer than the touch interval of multi-touch, it becomes impossible to distinguish a plurality of touches. It was.

According to one aspect of the present invention, it is provided with a touch panel in which a plurality of touch sensors are arranged two-dimensionally, and a touch detection unit that detects multi-touch based on a two-dimensional distribution of signals detected by the touch sensors. In the touch detection device, the touch detection unit estimates the number of touches from the first detection unit that detects the touch point and the number of touches from the maximum of the two-dimensional distribution of the signal, and the sum of the signals in the region around the touch points. A touch that has a second detection unit and searches for another touch candidate from the peripheral area when the number of touches detected by the first detection unit and the number of touches estimated by the second detection unit do not match. A detector is provided.

According to another aspect of the present invention, the touch panel includes a touch panel in which a plurality of touch sensors are arranged two-dimensionally, and a touch detection unit that detects multi-touch based on the two-dimensional distribution of signals detected by the touch sensors. A touch detection method used in a touch detection device, in which the number of touches is estimated from the first detection step of detecting the touch point and the number of touches from the maximum of the two-dimensional distribution of the signal and the sum of the signals in the region around the touch points. If the number of touches detected in the first detection step and the number of touches estimated in the second detection step do not match in the second detection step to be performed, a search step for searching other touch candidates from the peripheral area, and a search step. A touch detection method having the above is provided.

According to the present invention, it is possible to provide a touch detection device and a touch detection method capable of improving the resolution of detecting a plurality of touches at the same time.

It is a block diagram which conceptually shows the structure of the touch detection apparatus which concerns on 1st Embodiment. It is a figure which shows typically the example of the structure of the mutual capacitance detection type capacitance type touch panel. It is a figure which shows typically the example of the arrangement of the touch sensor in the touch detection apparatus which concerns on 1st Embodiment. It is a figure for demonstrating the search process of a touch candidate in the touch detection apparatus which concerns on 1st Embodiment. It is a flowchart which shows the touch detection method which concerns on 1st Embodiment. It is a figure for demonstrating the centroid calculation process in the touch detection method which concerns on 2nd Embodiment.

Hereinafter, preferred embodiments of the present invention will be described with reference to the drawings. The present invention is not limited to the following embodiments, and can be appropriately modified without departing from the gist thereof. In addition, those having the same or equivalent functions in each figure may be designated by the same reference numerals, and the description thereof may be omitted or simplified.

(First Embodiment)
Hereinafter, the touch detection device according to the first embodiment of the present invention will be described. FIG. 1 is a block diagram conceptually showing the configuration of the touch detection device according to the first embodiment. The touch detection device of the present embodiment includes a touch panel 1 and a control device 2. The touch detection device shown in FIG. 1 can be applied to, for example, a display device, a mobile terminal, or the like.

The touch panel 1 may be any as long as it can detect multi-touch in principle. For example, FIG. 2 schematically shows an example of the configuration of a mutual capacitance detection type capacitance type touch panel 1. As shown in FIG. 2, the mutual capacity detection type touch panel 1 has a plurality of row electrodes 101 and column electrodes 102 facing each other with an insulating layer interposed therebetween. Then, the touch is detected by measuring a signal based on the change in the capacitance of the touch sensor, which will be described later, provided at the intersection of the row electrode 101 and the column electrode 102. The touch panel 1 may be, for example, a self-capacity detection type in addition to the mutual capacity detection type.

The control device 2 is a semiconductor IC provided with a microprocessor and a memory for controlling the touch panel 1, and includes a touch detection unit 20, a drive unit 21, and a control unit 22. As shown in FIG. 2, the drive unit 21 applies a voltage to the touch sensor via the row electrode 101 and the column electrode 102 of the touch panel 1.

The touch detection unit 20 detects multi-touch by analyzing the two-dimensional distribution of the signal based on the change in the capacitance of the touch sensor provided at the intersection of the row electrode 101 and the column electrode 102. The touch detection unit 20 of the present embodiment is characterized by having a first detection unit 201 and a second detection unit 202. The specific operations of the first detection unit 201 and the second detection unit 202 will be described later with reference to FIG.

The control unit 22 controls the entire touch panel 1 in response to an input from the outside of the control device 2, and outputs the touch data detected by the touch detection unit 20 to the outside of the control device 2.

FIG. 3 is a diagram schematically showing an example of an arrangement of touch sensors 100 in the touch detection device according to the first embodiment. The touch panel 1 of the present embodiment has a plurality of touch sensors 100 arranged two-dimensionally.

3 (a) to 3 (c) show an example in which the single touch 110, the multi-touch 111 to 113, and the big touch 114 are performed on the arrangement of the touch sensors 100, respectively. Further, FIGS. 3 (d) to 3 (f) show two dimensions of the signal based on the change in the capacitance of the touch sensor 100 when the single touch 110, the multi-touch 111-113, and the big touch 114 are performed. The distribution is shown respectively.

Here, the single touch 110 is a case where the touch is performed by, for example, one finger. Further, the multi-touchs 111 to 113 are cases where, for example, touches are performed simultaneously by a plurality of fingers. The big touch 114 is, for example, a case where a touch is performed over a wide area with a thumb or the like having a large area, or a case where water droplets or the like adhere to the capacitive touch panel 1.

In addition, although only 33 touch sensors 100 of 6 rows × 6 columns (or 5 columns) are shown in FIGS. 3 (a) to 3 (c), there are more general touch panels 1. Has a touch sensor 100. Further, in FIGS. 3A to 3C, the shape of the touch sensor 100 is a cross shape, but the shape of the touch sensor 100 is a cross shape in which the length in the row direction and the length in the column direction are equal to each other. It may be present, or it may be in the shape of a rectangle, a square, or the like. By forming the touch sensor 100 into a cross shape or a cross shape, the number of adjacent touch sensors 100 can be increased, so that the accuracy in the calculation of the center of gravity, which will be described later, can be improved.

The single touch 110, the multi-touch 111-113, and the big touch 114 are each characterized by the acquired signal distribution, and the two-dimensional distribution of the signal as shown in FIGS. 3 (d) to 3 (f) is analyzed. It is possible to discriminate. For example, as shown in FIG. 3 (f), the big touch 114 can be discriminated by having a maximum signal that exceeds a predetermined threshold value (200 in FIG. 3 (f)).

Next, a method for detecting the single touch 110 shown in FIG. 3A will be described. As described above, the touch detection unit 20 of the present embodiment is characterized by having a first detection unit 201 and a second detection unit 202. First, the first detection unit 201 detects the position coordinates touched by the user (hereinafter, "touch points") and the total number of touch points (hereinafter, "touch numbers") from the maximum of the signal distribution. For example, in FIG. 3D, the maximum (= 90) exceeding a predetermined first threshold value (= 10) in the signal distribution is detected as the single touch 110.

Further, the first detection unit 201 adds (labels) a unique ID to the detected single touch 110. Then, the labeling region 120 around the single touch 110 is set as a region that characterizes the signal distribution of the single touch 110. For example, in FIG. 3D, the regions of the six touch sensors 100 adjacent to the maximum of the single touch 110 are set as the labeling regions 120.

The labeling area 120 set here is also used as a center of gravity calculation area for calculating the center of gravity in the subsequent center of gravity calculation process. The range of the labeling area 120 can be changed according to the magnitude of the detected maximum value, etc. However, if the labeling area 120 is made too wide, it takes time to calculate the center of gravity later, and the touch panel 1 has a touch panel 1. Since the response will be slow, it can be adjusted appropriately according to the application.

On the other hand, the second detection unit 202 estimates the number of touches from the total number of signals in the labeling region 120 set by the first detection unit 201. For example, in FIG. 3D, the sum of signals (= 127) in the labeling region 120 is divided by a predetermined second threshold value (= 80), and the value (127/80 = 1.59) is rounded down to the nearest whole number. The value (= 1) is estimated as the number of touches. The touch detection unit 20 determines that the single touch 110 has been performed because the number of touches detected by the first detection unit 201 and the number of touches estimated by the second detection unit 202 both match with 1. .. The peripheral region of the single touch 110 for which the total sum is calculated by the second detection unit 202 may be a region different from the labeling region 120. For example, the sum may be calculated in the region where some signal (greater than or equal to a predetermined threshold) is detected.

Next, the detection method of the multi-touch 111 to 113 shown in FIG. 3B will be described. First, the first detection unit 201 detects the touch points and the number of touches touched by the user from the maximum of the signal distribution. For example, in FIG. 3E, the maximum (= 81, 72) exceeding a predetermined first threshold value (= 10) in the signal distribution is detected as multi-touch 111 and 112, respectively.

Further, the first detection unit 201 adds (labels) a unique ID to the detected multi-touch 111, 112. Then, the labeling regions 121 and 122 around the multi-touch 111 and 112 are set as regions that characterize the signal distribution of the multi-touch 111 and 112, respectively. For example, in FIG. 3E, the regions of the six touch sensors 100 adjacent to the maximum of the multi-touch 111 are set as the labeling regions 121. Further, the regions of the six touch sensors 100 adjacent to the maximum of the multi-touch 112 are set as the labeling regions 122.

Here, in FIG. 3E, the first detection unit 201 has not been able to detect the multi-touch 113, and has detected that the number of touches is 2. This is because the multi-touch 112 and 113 are too close to each other in FIG. 3B, and the arrangement interval of the touch sensors 100 is equal to or greater than the touch interval of the multi-touch 112 and 113. .. As a result, in FIG. 3E, the signal of the multi-touch 112 is superimposed on the signal of the multi-touch 113, and the maximum of the multi-touch 113 disappears. Further, in FIG. 3E, the signal of the multi-touch 113 is dispersed in the two touch sensors 100, which is also one of the reasons why the first detection unit 201 cannot detect the maximum of the multi-touch 113. There is.

Therefore, the second detection unit 202 estimates the number of touches from the sum of the signals in the labeling areas 121 and 122 set by the first detection unit 201. For example, in FIG. 3 (e), the sum of signals (= 292) in the labeling regions 121 and 122 is divided by a predetermined second threshold value (= 80), and the value (292/80 = 3.65) is below the decimal point. The value (= 3) rounded down is estimated as the number of touches. The peripheral regions of the multi-touch 111 and 112 for which the total sum is calculated by the second detection unit 202 may be different from the labeling regions 121 and 122. For example, the sum may be calculated in the region where some signal (greater than or equal to a predetermined threshold) is detected.

The touch detection unit 20 does not match the number of touches (= 2) detected by the first detection unit 201 with the number of touches (= 3) estimated by the second detection unit 202, so that the first detection unit 20 detects it. It is determined that there is another touch point that was not detected by unit 201. Then, another touch candidate is searched from within the labeling areas 121 and 122. The search process for the touch candidate will be described later with reference to FIG.

As described above, in the present embodiment, the first detection unit 201 that detects the touch point from the maximum of the signal distribution and the second detection unit 202 that estimates the number of touches from the total sum of the signals are used in combination. Thereby, even when the arrangement interval of the touch sensor 100 is equal to or longer than the touch interval of the multi-touch 112 and 113, a plurality of touches can be distinguished.

Further, when the big touch 114 shown in FIG. 3C is also discriminated, the big touch 114 can be discriminated more accurately by using the second detection unit 202 together. For example, in FIG. 3 (f), the total sum of the signals in the labeling region 124 around the big touch 114 is 2631, which is very large. Therefore, the maximum in which the total sum of the signals in the labeling region 124 exceeds a predetermined third threshold value (2000 in FIG. 3 (f)) can be determined as the big touch 114.

The first threshold value used by the first detection unit 201 and the second threshold value used by the second detection unit 202 may be set for each position coordinate of the touch sensor 100. As a result, even when the touch sensor 100 arranged at the end of the touch panel 1 and the touch sensor 100 arranged at the center of the touch panel 1 have different sensitivity levels, the touch can be accurately detected.

FIG. 4 is a diagram for explaining a touch candidate search process in the touch detection device according to the first embodiment. FIGS. 4 (b) and 4 (e) shown on the left side are the same views as those of FIGS. 3 (b) and 3 (e) above. On the other hand, FIGS. 4 (g) and 4 (h) show an example of a process of searching for another touch candidate when the number of touches does not match between the first detection unit 201 and the second detection unit 202. Shown.

In FIG. 4 (g), the largest signal (= 39) in the labeling regions 121 and 122 except for the multi-touch 111 and 112 is searched as a touch candidate. This touch candidate is the multi-touch 113 which is too close to the multi-touch 112 and could not be detected by the first detection unit 201. In the present embodiment, by using the first detection unit 201 and the second detection unit 202 together in this way, it is possible to distinguish between the multi-touch 112 and 113 even when the touch interval is small. In the search of touch candidates, it is possible to perform more advanced search processing by expanding the search range, but if the algorithm is too complicated, the search processing will take time and the response of the touch panel 1 will be slow. Therefore, it can be adjusted as appropriate according to the application.

After that, the first detection unit 201 again detects the touch point and the number of touches from the maximum value of the signal distribution, including the searched touch candidates. Then, in FIG. 4 (h), a unique ID is added (labeled) to the detected multi-touch 111 to 113, and the labeling areas 121 to 123 are designated as regions for characterizing the signal distribution of the multi-touch 111 to 113, respectively. Set.

As a result, the number of touches detected by the first detection unit 201 and the number of touches estimated by the second detection unit 202 both match at 3, so that the touch detection unit 20 has three multi-touches 111 to 113. Determine that it was done. After that, the touch detection unit 20 calculates the center of gravity of the detected multi-touch 111 to 113. A specific method for calculating the center of gravity will be described later with reference to FIG. 6 in the second embodiment.

FIG. 5 is a flowchart showing a touch detection method according to the first embodiment. FIG. 5 shows a flowchart of the detection methods of the multi-touch 111 to 113 described so far.

First, in step S01, the drive unit 21 applies a voltage to the touch sensor 100 via the row electrode 101 and the column electrode 102 of the touch panel 1. The touch detection unit 20 AD-converts a signal based on a change in the capacitance of the touch sensor 100 provided at the intersection of the row electrode 101 and the column electrode 102, and acquires a two-dimensional distribution thereof.

Next, in step S02, the touch detection unit 20 analyzes the two-dimensional distribution of the signal acquired in step S01 and sets a position calculation area for detecting the touch point. For this position calculation area, for example, an area in which some signal (at least a predetermined threshold value) is detected in step S01 is set. Subsequent processing is performed in this position calculation area.

In step S03, the first detection unit 201 detects the touch point and the number of touches from the maximum of the signal distribution. On the other hand, in step S04, the second detection unit 202 estimates the number of touches from the sum of the signals in the region around the touch point. Here, either step S03 or step S04 may be performed first. When the estimation process by the second detection unit 202 is performed after the detection process by the first detection unit 201, the sum total of the signals is calculated with the labeling areas 121 and 122 shown in FIG. 4 as peripheral areas. On the other hand, when the estimation process by the second detection unit 202 is performed before the detection process by the first detection unit 201, the total sum of the signals is calculated with the position calculation area set in step S02 as the peripheral area.

In step S05, the touch detection unit 20 determines whether or not the number of touches detected by the first detection unit 201 and the number of touches estimated by the second detection unit 202 match. If the numbers of touches match (YES), the process proceeds to step S06, the coordinate correction process including the calculation of the center of gravity of the detected multi-touchs 111 to 113 is performed, and the touch detection process ends. On the other hand, if the touch numbers do not match (NO), the process proceeds to step S07.

In step S07, the touch detection unit 20 determines that there is another touch point that has not been detected by the first detection unit 201, and from the labeling areas 121 and 122 set by the first detection unit 201 to other touch points. Search for touch candidates. This search process is as described with reference to FIG. 4, for example.

After that, returning to step S03, the first detection unit 201 again detects the touch point and the number of touches from the maximum of the signal distribution, including the searched touch candidates. Then, in step S05, the touch detection unit 20 repeats steps S07 to S05 until the number of touches detected by the first detection unit 201 and the number of touches estimated by the second detection unit 202 match. When the number of touches matches, the process proceeds to step S06, the coordinate correction process including the calculation of the center of gravity of the detected multi-touchs 111 to 113 is performed, and the touch detection process ends.

As described above, in the present embodiment, the first detection unit that detects the touch point and the number of touches from the maximum of the two-dimensional distribution of the signal detected by the touch sensor, and the number of touches from the sum of the signals in the region around the touch points. It has a second detection unit that estimates. Then, when the number of touches detected by the first detection unit and the number of touches estimated by the second detection unit do not match, another touch candidate is searched from the peripheral area.

According to such a configuration, even when the arrangement interval of the touch sensors 100 is equal to or greater than the touch interval, or when the touches are distributed among the two touch sensors 100, a plurality of touches are made. Can be distinguished. In particular, in the present embodiment, in an in-cell type or on-cell type touch detection device used for a display that has been increasing in size in recent years, it is possible to improve the resolution of simultaneously detecting a plurality of touches while suppressing power consumption and cost.

(Second Embodiment)
Hereinafter, the touch detection device according to the second embodiment of the present invention will be described. In this embodiment, a method of performing the coordinate correction process in step S06 shown in FIG. 5 with higher accuracy will be described. FIG. 6 is a diagram for explaining the center of gravity calculation process in the touch detection method according to the second embodiment.

FIG. 6 (h) shown on the far left is the same diagram as in FIG. 4 (h) above. On the other hand, FIGS. 6 (i) to 6 (m) show an example of the center of gravity calculation process for the multi-touch 111 to 113 detected by using the method described in the first embodiment. In the following, the description will be focused on the points different from the first embodiment, and the description overlapping with the first embodiment may be omitted or simplified.

First, a method of calculating the center of gravity of the first touch 111 will be described with reference to FIG. 6 (i). There are no other second touch 112 or third touch 113 in the labeling area 121 of the first touch 111. That is, since the signal distribution of the first touch 111 is less affected by the signals of the other second touch 112 and the third touch 113, a normal center of gravity calculation method can be used. Specifically, the touch detection unit 20 _{calculates the center of gravity x G} of the first touch 111 by the following equation (1).
x _G = Σ (s (x _ij ) · x _ij ) / _{Σ s (x ij} ) (1)

Here, the center of gravity x _G and the position coordinates x _ij are vectors. Signal s _{(x ij)} is the detected signal value with the touch sensor 100 which is disposed at a position coordinate _{x ij,} subscript coordinates _{x ij} is the row number of the row electrodes 101 in which the touch sensor 100 is arranged It represents i and the row number j of the row electrode 102. The sum of Σ is performed for _{all the position coordinates x ij in the labeling region 121.}

Next, the method of calculating the center of gravity of the second touch 112 will be described with reference to FIGS. 6 (j) and 6 (l). Within the labeling area 122 of the second touch 112, there is another third touch 113. That is, since the signal distribution of the second touch 112 is greatly influenced by the signals of the other third touch 113, if the same center of gravity calculation as that of the first touch 111 is performed as it is, the center of gravity of the second touch 112 becomes the second. The calculation is largely biased toward the touch 113 side of 3.

Therefore, in the present embodiment, when another third touch 113 exists in the center of gravity calculation area of the second touch 112, a correction for lowering the signal level in the center of gravity calculation area of the other third touch 113 is performed. After that, the center of gravity of the above equation (1) is calculated. For example, in FIG. 6 (l), after correcting the signal value in the labeling region 123 to 1/16, the center of gravity x _G of the second touch 112 is calculated by the above equation (1). As a result, the influence of the adjacent third touch 113 can be mitigated, and the center of gravity of the second touch 112 can be calculated accurately.

Similarly, as shown in FIG. 6 (m), the method of calculating the center of gravity of the third touch 113 is also the above equation (1) after the correction for lowering the signal level in the center of gravity calculation region of the second touch 112 is performed. Calculate the center of gravity of. As a result, the influence of the adjacent second touch 112 can be mitigated, and the center of gravity of the third touch 113 can be calculated accurately.

As described above, when another touch point exists in the detected center of gravity calculation area of the touch point, the touch detection unit of the present embodiment sets a predetermined value for a signal in the center of gravity calculation area of the other touch point. After dividing, calculate the centroid of the touchpoint. According to such a configuration, the resolution for detecting a plurality of touches at the same time can be further improved.

In the above description, the labeling areas 121 to 123 are used as the center of gravity calculation area for calculating the center of gravity, but an area different from the labeling areas 121 to 123 may be set as the center of gravity calculation area. For example, a region within a certain distance range from the detected touch coordinates of the multi-touch 111 to 113 may be set as the center of gravity calculation region.

(Other embodiments)
The above-described embodiments are merely examples of embodiment in carrying out the present invention, and the technical scope of the present invention should not be construed in a limited manner by these. That is, the present invention can be implemented in various forms without departing from the technical idea or its main features.

1: Touch panel 2: Control device 20: Touch detection unit 201: First detection unit 202: Second detection unit 21: Drive unit 22: Control unit 100: Touch sensor 111-113: Multi-touch 120 to 124: Labeling area (Center of gravity calculation area)

Claims (13)

A touch detection device including a touch panel in which a plurality of touch sensors are arranged in a two-dimensional manner and a touch detection unit that detects multi-touch based on a two-dimensional distribution of signals detected by the touch sensors. In the touch detector
A first detection unit that detects the touch point and the number of touches from the maximum of the two-dimensional distribution of the signal, and
A second detector that estimates the number of touches from the sum of the signals in the area around the touch point,
A touch detection device that searches for other touch candidates from the peripheral area when the number of touches detected by the first detection unit and the number of touches estimated by the second detection unit do not match. The touch panel has a plurality of row electrodes and column electrodes facing each other with an insulating layer interposed therebetween.
The touch detection device according to claim 1, wherein the touch detection unit acquires a two-dimensional distribution of the signal based on a change in the capacitance of the touch sensor provided at an intersection of the row electrode and the column electrode. The first detection unit detects the maximum that exceeds a predetermined first threshold value in the two-dimensional distribution of the signal as the touch points, and detects the total number of the touch points as the number of touches. 2. The touch detection device according to 2. The touch detection device according to claim 3, wherein the second detection unit estimates a value obtained by dividing the sum by a predetermined second threshold value as the number of touches. The touch detection device according to claim 4, wherein the first threshold value or the second threshold value is set for each position coordinate of the touch sensor. The touch detection device according to claim 5, wherein the first detection unit determines a maximum in which the sum exceeds a predetermined third threshold value in the two-dimensional distribution of the signal as a big touch. The touch detection device according to any one of claims 1 to 6, wherein the touch detection unit searches for the largest signal in the peripheral region as the touch candidate. When another touch point exists in the center of gravity calculation area of the touch point, the touch detection unit divides the signal in the center of gravity calculation area of the other touch point by a predetermined value, and then divides the signal by a predetermined value. The touch detection device according to any one of claims 1 to 7, which calculates the center of gravity of the touch point. The touch detection device according to any one of claims 1 to 8, wherein the touch sensor has a cross shape. The touch detection device according to any one of claims 1 to 9, further comprising an in-cell type or on-cell type touch panel used for a display. Touch detection used in a touch detection device including a touch panel in which a plurality of touch sensors are arranged in a two-dimensional manner and a touch detection unit that detects multi-touch based on a two-dimensional distribution of signals detected by the touch sensors. It ’s a method,
The first detection step of detecting the touch point and the number of touches from the maximum of the two-dimensional distribution of the signal, and
A second detection step of estimating the number of touches from the sum of the signals in the region around the touch point, and
If the number of touches detected in the first detection step and the number of touches estimated in the second detection step do not match, a search step of searching for another touch candidate from the peripheral area and a search step
Touch detection method having. The touch detection method according to claim 11, wherein the second detection step is performed before the first detection step. The touch detection method according to claim 11, wherein after searching for the touch candidate in the search step, the first detection step is performed again including the touch candidate.

Images