CN-121523569-B - Touch screen false touch prevention control method and system based on multi-sensor fusion

CN121523569BCN 121523569 BCN121523569 BCN 121523569BCN-121523569-B

Abstract

The invention relates to the technical field of man-machine interaction and discloses a touch screen false touch prevention control method and a touch screen false touch prevention control system based on multi-sensor fusion, wherein touch data and inertial data are acquired by the method, and comprehensive analysis is carried out from three dimensions, namely, the divergence characteristics of a gradient vector field are analyzed based on an original capacitance matrix, and an electric field characteristic index representing a proximity mode is output; calculating the amplitude square coherence function of the integral vibration signal and the contact micro-displacement signal of the equipment in a preset physiological frequency band, outputting a coupling coefficient representing a physical connection state, counting the directional state probability distribution of a contact moving track, calculating shannon entropy, and outputting a track entropy value representing the order of operation intention. The method further calculates the false touch confidence of the contact according to the electric field characteristic index, the coupling coefficient and the track entropy value, and generates a dynamic suppression mask based on the confidence so as to intercept false touch events with high confidence at the bottom layer. The invention can obviously improve the identification accuracy and the robustness through multi-dimensional feature fusion.

Inventors

LI KAIHUA
GONG BAOYU
HU YING
LI JINYUAN
Hu Zuning
PENG YONG
LI YUQUN
LIN QINGFANG

Assignees

江西权智电子科技有限公司

Dates

Publication Date: 20260508
Application Date: 20251121

Claims (6)

1. The touch screen false touch prevention control method based on multi-sensor fusion is characterized by comprising the following steps of: Acquiring touch data and inertial data; Calculating the divergence characteristic of a gradient vector field of a contact area based on an original capacitance matrix in the touch data, and outputting an electric field characteristic index representing a proximity mode; Based on the contact coordinate data and the inertia data in the touch data, calculating an amplitude square coherence function of an integral vibration signal and a contact micro-displacement signal of the equipment in a preset physiological frequency band, and outputting a coupling coefficient representing the physical connection state of the contact and the equipment; Based on the contact coordinate data, counting the directional state probability distribution of the contact movement track, calculating shannon entropy, and outputting a track entropy value representing the order of the operation intention; calculating the false touch confidence of the contact according to the electric field characteristic index, the coupling coefficient and the track entropy value; generating a dynamic inhibition mask according to the false touch confidence, and intercepting a touch event in the dynamic inhibition mask; The step of outputting the electric field characteristic index representing the approaching mode comprises the following steps: Calculating the capacitance gradient vector of each induction node based on the original capacitance matrix; calculating the divergence characteristic of each sensing node based on the original capacitance matrix; In the contact area, synthesizing the module value of the divergence characteristic and the capacitance gradient vector, and calculating to obtain the electric field characteristic index; the step of outputting a coupling coefficient representing the physical connection state of the contact and the device comprises the following steps: under the condition that the energy of the integral vibration signal of the equipment is higher than a preset static noise threshold value, integrating the amplitude square coherence function in the preset physiological frequency band to obtain the coupling coefficient; the amplitude square coherence function is calculated based on the self-power spectral density of the whole vibration signal of the equipment, the self-power spectral density of the contact micro-displacement signal and the cross-power spectral density between the two signals; the step of generating the dynamic suppression mask specifically includes: When the false touch confidence is higher than a preset inhibition threshold, identifying an original capacitance profile of the contact; And enveloping the original capacitance profile by adopting a convex hull algorithm to generate the dynamic suppression mask.
2. The method for controlling a touch screen to prevent false touch based on multi-sensor fusion according to claim 1, wherein the step of outputting a trajectory entropy value representing the order of operation intention comprises: Encoding the movement track of the contact as a direction chain code sequence; counting the occurrence probability of each direction state in the direction chain code sequence in a preset time window; and calculating the track entropy value according to the occurrence probability.
3. The touch screen false touch prevention control method based on multi-sensor fusion according to claim 2, further comprising, after calculating the track entropy value: Converting the track entropy value into an intention checking coefficient through a preset S-shaped mapping function; the false touch confidence is calculated based on the intent verification coefficient.
4. The method for preventing false touch on a touch screen based on multi-sensor fusion according to claim 1, wherein the step of calculating the false touch confidence of the touch point specifically comprises the following steps: And obtaining the false touch confidence coefficient by carrying out weighted linear combination on the electric field characteristic index, the coupling coefficient and an index which is inversely related to the track entropy value.
5. The method for preventing false touch on a touch screen based on multi-sensor fusion according to claim 1, wherein the step of intercepting the touch event located in the dynamic suppression mask further comprises: and when the touch event is a moving event and the moving track of the touch event continuously moves out of the dynamic inhibition mask, and after the moving distance reaches a preset separation threshold value, releasing interception of the touch event.
6. A touch screen anti-false touch control system based on multi-sensor fusion, for executing the touch screen anti-false touch control method based on multi-sensor fusion as claimed in any one of claims 1 to 5, comprising a central processing unit, wherein the central processing unit is internally configured with: The data synchronous acquisition module is used for acquiring touch data and inertial data; The electric field characteristic analysis module is used for calculating the divergence characteristic of the gradient vector field of the contact area based on the original capacitance matrix in the touch data and outputting an electric field characteristic index representing the approaching mode; The biomechanical coupling analysis module is used for calculating an amplitude square coherence function of the whole vibration signal and the contact micro-displacement signal of the equipment in a preset physiological frequency band based on contact coordinate data and inertia data in the touch data, and outputting a coupling coefficient representing the physical connection state of the contact and the equipment; The track entropy checking module is used for counting the directional state probability distribution of the contact moving track based on the contact coordinate data, calculating shannon entropy and outputting a track entropy value representing the order of the operation intention; And the multidimensional decision control module is used for calculating the false touch confidence coefficient of the contact according to the electric field characteristic index, the coupling coefficient and the track entropy value, generating a dynamic inhibition mask according to the false touch confidence coefficient, and sending an instruction for intercepting a touch event in the dynamic inhibition mask to a bottom layer driver.

Description

Touch screen false touch prevention control method and system based on multi-sensor fusion Technical Field The invention relates to the technical field of man-machine interaction, in particular to a touch screen false touch prevention control method and system based on multi-sensor fusion. Background Touch screens have become the dominant input interface for electronic devices such as smartphones, tablet computers, and vehicle terminals. The quality of the user interaction experience depends largely on the accuracy of the touch screen's response to valid operating instructions, as well as the ability to inhibit inadvertent contact. However, in actual use, the false touch problem often affects the user experience. This phenomenon occurs frequently, especially when the user holds the device, and the non-operating parts such as the palm and knuckles may inadvertently touch the surface of the screen, which is more and more the problem, especially in the design trend of the increasingly narrowing screen borders. To solve this problem, the existing false touch prevention technology generally relies on a single physical feature to make the judgment. For example, one common approach is to distinguish based on the size of the contact area, i.e., assuming that the contact with an area exceeding a certain threshold is a false touch by the palm or the like. Although this method can filter a part of the apparent large-area contact, it has limited discrimination capability. It is difficult to distinguish between intended flat finger operation and unintended palm root contact with similar press areas, and it is also possible that a valid multi-finger gesture may be erroneously recognized as a false touch, or that a false touch from a smaller area such as a knuckle may not be recognized. In order to make up for the defects, some schemes increase judging bases such as contact time length or static property of contact positions. However, these methods also have limitations. The user may have a long and relatively fixed contact point on his hand while holding the device, similar to the behavior of a user's intent to press a screen icon for a long time, resulting in indistinguishability. At the same time, small movements of the gripping hand on the device may also be erroneously interpreted as intentional sliding operations. The fundamental limitation of the prior art in false touch prevention judgment is that the analysis dimension is limited to two-dimensional plane data generated by the touch screen, and depth insight of a background generated by a touch event is lacking. In particular, these techniques are not effective in distinguishing between the physical source of the contact, whether it is from a finger that is separate from the motion of the device and actively operating, or from a grip portion that is mechanically coupled to the body of the device and moves with the device. This inherent ambiguity makes it difficult for the prior art to have to make a compromise between recognition sensitivity and accuracy, while avoiding loss of effective operation and interference of false touch operation. Disclosure of Invention The invention aims to solve the technical problems that when the conventional touch screen anti-false touch technology distinguishes the intentional operation and unintentional touch under the complex scene, the problems of single recognition dimension and low accuracy exist, and the large-area and long-time static or dynamic false touch generated by the palm root, the knuckle and other parts when the device is held is difficult to accurately recognize by judging only depending on single physical characteristics such as contact area or contact time length, so that the user experience is affected. In order to solve the technical problems, the invention provides a touch screen false touch prevention control method and system based on multi-sensor fusion. The invention provides a touch screen false touch prevention control method based on multi-sensor fusion, which is used for establishing a multi-dimensional comprehensive decision model by integrating electric field physical characteristics, biomechanical coupling characteristics and time-space track intention characteristics so as to realize accurate judgment of touch event properties. In one embodiment, the method first obtains touch data and inertial data. The touch data comprise an original capacitance matrix and contact coordinate data, and the inertia data comprise acceleration and angular velocity information of the device. In order to distinguish the contact properties from the initial form of physical contact, the method calculates the divergence characteristic of the gradient vector field of the contact area based on the original capacitance matrix in the touch data, and outputs an electric field characteristic index representing the approaching mode. Specifically, by calculating the capacitance gradient vector and the divergence characteristic of