CN219266985U

CN219266985U - Texture recognition sensing device based on optical fiber bionic skin

Info

Publication number: CN219266985U
Application number: CN202320867290.4U
Authority: CN
Inventors: 翁俊杰
Original assignee: Hunan Wanwei Zhigan Technology Co ltd
Current assignee: Hunan Wanwei Zhigan Technology Co ltd
Priority date: 2023-04-18
Filing date: 2023-04-18
Publication date: 2023-06-27
Anticipated expiration: 2033-04-18

Abstract

The utility model discloses a texture recognition sensing device based on optical fiber bionic skin, which comprises: the optical fiber bionic skin comprises a light source, optical fiber bionic skin, a photoelectric detector, a digital acquisition card, a computer and a signal processing device, wherein the two ends of the optical fiber bionic skin are respectively connected with the light source and the photoelectric detector, the photoelectric detector is connected with the digital acquisition card, the digital acquisition card is connected with the computer, and the computer is connected with the signal processing device. The texture recognition sensing device based on the optical fiber bionic skin transmits signals to the digital acquisition card through the micro-nano optical fiber bionic neuron, the digital acquisition card transmits data to a computer, and recognizes the surface texture characteristics of a target object through the signal processing device.

Description

Texture recognition sensing device based on optical fiber bionic skin

Technical Field

The utility model relates to the technical field of skin texture recognition, in particular to a texture recognition sensing device based on optical fiber bionic skin.

Background

Touch is one of five human senses, and is important for human interaction with the external environment. The touch is mainly derived from human finger skin, which has the most effective sensory system and sensory touch pattern (such as touch, pressure, vibration, warmth, coldness, pain), and the spatiotemporal perception of externally applied stimulus by skin sensory receptors. Spatiotemporal tactile signals transmitted by nerve afferent to the somatosensory cortex are then encoded as voltage peaks of action potentials and then transmitted to the brain. The brain will then comprehensively identify the type and intensity of the tactile stimulus. It is noted that the brain can actively memorize the characteristics of the stimulus signal and the associated tactile events, for example, the blind can learn braille (a character with a special surface structure) by repeatedly performing touch-memory, and the information attached to braille can be obtained through touch. However, human subjective haptic sensations are related not only to skin deformation and vibration, but also to other psychological factors such as memory, personality, expectations, and the like.

In recent years, optical fiber-based tactile sensors have been favored by researchers because of their high sensitivity, corrosion resistance, electromagnetic interference resistance, and no metallic material composition, as compared to the relatively mature electrical tactile sensors currently developed. Instead, micro-nano optical fibers are widely used in sensing fields including refractive index sensing, magnetic field sensing, ocean sensing and micro-force sensing, as an optical waveguide with a radius between (less than or near) sub-wavelengths, due to their high sensitivity, extreme flexibility and configurability. However, tactile sensors based on micro-nano optical fibers for touch force and texture sensing, as well as surface texture recognition in combination with neural network algorithms, are still rarely reported. Therefore, the bionic skin with stable sensing and identifying capability is of great significance for the development of the fields of next-generation medical care, robots, human-computer interaction and the like as the corresponding of the skin.

Disclosure of Invention

The utility model aims to provide a texture recognition sensing device based on optical fiber bionic skin, which aims to solve the problem that the prior art cannot quantitatively analyze the texture of a fabric. The sensor has the advantages of compact structure and simple sensing system, and can be used for solving the problem of texture digital information of the surface of a target object.

In order to achieve the above purpose, the present utility model provides the following technical solutions: a texture recognition sensing device based on optical fiber bionic skin, comprising: the optical fiber bionic skin comprises a light source, optical fiber bionic skin, a photoelectric detector, a digital acquisition card, a computer and a signal processing device, wherein the two ends of the optical fiber bionic skin are respectively connected with the light source and the photoelectric detector, the photoelectric detector is connected with the digital acquisition card, the digital acquisition card is connected with the computer, and the computer is connected with the signal processing device.

Preferably, the light source is used for inputting light to the optical fiber bionic skin, and the photoelectric detector is used for receiving and detecting the output light.

Preferably, the signal processing device is used for identifying texture features of the surface of the target object.

Preferably, the optical fiber bionic skin comprises a bionic fingerprint structure, an upper bionic skin layer, a side bionic skin layer and a substrate, wherein the surface of the upper bionic skin layer is provided with the fingerprint-shaped bionic fingerprint structure, and the side bionic skin layer is arranged between the upper bionic skin layer and the substrate.

Preferably, the side bionic dermis layer comprises an encapsulation layer and micro-nano optical fiber bionic neurons, and the micro-nano optical fiber bionic neurons penetrate through two ends of the encapsulation layer.

Preferably, the micro-nano optical fiber bionic neuron comprises a single-mode fiber region, a taper transition region of the micro-nano optical fiber and a uniform waist region of the micro-nano optical fiber, wherein the taper transition region of the micro-nano optical fiber is integrally connected to two ends of the uniform waist region of the micro-nano optical fiber, and the other end of the taper transition region of the micro-nano optical fiber is integrally connected with the single-mode fiber region.

Compared with the prior art, the utility model has the beneficial effects that: this texture recognition sensing device based on bionical skin of optic fibre can be after the object laminating is on bionical fingerprint structure, the pressure of laminating is conducted to the bionical neuron of little nanometer optic fibre through last bionical epidermis layer, thereby make the bionical neuron of little nanometer optic fibre produce little bending, afterwards the bionical neuron of little nanometer optic fibre transmits the signal for digital acquisition card, digital acquisition card transmits data, and transmit the computer, and discern the surface texture characteristic of target object through signal processing device, this scheme simple structure, compactness, can convert target object surface texture into digital information, afterwards can carry out quantitative analysis to digital information, thereby discern the object.

Drawings

FIG. 1 is a schematic view of an optical path of a texture recognition sensing device based on fiber-optic biomimetic skin;

FIG. 2 is a schematic diagram of the structure of a fiber biomimetic skin;

fig. 3 is a schematic structural diagram of a micro-nano optical fiber biomimetic neuron.

In the figure: 1. a light source; 2. optical fiber bionic skin; 3. a photodetector; 4. a digital acquisition card; 5. a computer; 6. a signal processing device; 7. a bionic fingerprint structure; 8. an upper bionic epidermis layer; 9. a side bionic dermis layer; 10. an encapsulation layer; 11. micro-nano optical fiber bionic neurons; 1101. a single mode optical fiber region; 1102. a tapered transition region of the micro-nano optical fiber; 1103. a uniform waist region of the micro-nano optical fiber; 12. a substrate.

Detailed Description

The following description of the embodiments of the present utility model will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present utility model, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the utility model without making any inventive effort, are intended to be within the scope of the utility model.

Referring to fig. 1-3, the present utility model provides a technical solution: a texture recognition sensing device based on optical fiber bionic skin, comprising: the optical fiber bionic skin 2 is connected with the light source 1 and the photoelectric detector 3 at two ends of the optical fiber bionic skin 2 respectively, the light source 1 is used for inputting light to the optical fiber bionic skin 2, the optical fiber bionic skin 2 comprises a bionic fingerprint structure 7, an upper bionic skin layer 8, a side bionic skin layer 9 and a substrate 12, the surface of the upper bionic skin layer 8 is provided with the fingerprint-shaped bionic fingerprint structure 7, the hardness of the upper bionic skin layer 8 is higher than that of the side bionic skin layer 9, the tactile signal is prevented from being absorbed by materials before being transmitted to the side bionic skin layer 9, so that the sensitivity of the optical fiber bionic skin 2 is ensured, the bionic fingerprint structure 7 imitates the external skin and fingerprints of human beings and is used for increasing friction coefficients so as to achieve the effect of amplifying tactile vibration caused by friction, a side bionic dermis layer 9 is arranged between the upper bionic epidermis layer 8 and the substrate 12, the substrate 12 imitates bones in human skin and is used for supporting and protecting bionic optical skin from breakage and damage, the side bionic dermis layer 9 comprises a packaging layer 10 and micro-nano optical fiber bionic neurons 11, the micro-nano optical fiber bionic neurons 11 play a role in sensing and transmitting external stimulus and convert the external stimulus into optical signals with tactile information, the micro-nano optical fiber bionic neurons 11 penetrate through two ends of the packaging layer 10, the micro-nano optical fiber bionic neurons 11 comprise a single-mode optical fiber region 1101, a conical transition region 1102 of the micro-nano optical fiber and a uniform waist region 1103 of the micro-nano optical fiber, the diameter of the uniform waist region 503 is 2 mu m, the length is 5mm, the two ends of the uniform waist region 1103 of the micro-nano optical fiber are integrally connected with the conical transition regions 1102 of the micro-nano optical fiber, the other end of the conical transition region 1102 of the micro-nano optical fiber is integrally connected with a single mode fiber region 1101, the photoelectric detector 3 is used for receiving and detecting output light, the photoelectric detector 3 is connected with the digital acquisition card 4, the digital acquisition card 4 is connected with the computer 5, tactile signals on the computer 5 are acquired through the digital acquisition card 4, a data set containing a certain number of samples is constructed, a deep learning model is adopted to classify and predict tactile signals related to texture features, the computer 5 is connected with the signal processing device 6, and the signal processing device 6 is used for identifying the texture features on the surface of a target object.

Embodiments are described below: the tactile signals on the computer 5 are acquired through the digital acquisition card 4, a data set containing a certain number of samples is constructed, and the deep learning model is adopted to realize classification and prediction of the tactile signals related to the texture features. The original data is a waveform curve containing 2200 sampling points, the original data set is divided into signal data with the amplitude of more than about 200 time points of each sample through downsampling, in addition, the sample size of the data set is further expanded by adopting a data enhancement method, and then the complete data set is divided into a training set and a testing set to respectively train and test the performance of the neural network. The deep learning model selects a fully-connected neural network comprising a plurality of hidden layers, the accuracy of the neural network for sample classification in a training set and a testing set is gradually improved along with the training, the neural network can show accuracy similar to the training set for the testing set, and the neural network is proved to show stronger generalization capability on the testing set. The final result shows that the recognition rate of signals generated by the neural network model after training for 8 different fabrics can reach more than 90%.

When the optical fiber bionic skin 2 is in contact with an object, vertical pressure is conducted to the micro-nano optical fiber bionic neuron 11 through the upper bionic skin layer 8 to cause micro-bending of the micro-nano optical fiber bionic neuron 11, so that bending loss of the optical fiber is increased, and output light intensity is reduced, so that pressure sensing can be realized by monitoring change of the output light intensity. The optical fiber bionic skin 2 can detect a transverse shearing force in addition to a vertical pressure. The bionic fingerprint structure 7 on the upper bionic epidermis layer 8 decomposes the shearing force into vertical downward pressure, and the pressure is transmitted to the micro-nano optical fiber bionic neuron 11, so that the sensing of the shearing force is realized.

When the optical fiber bionic skin 2 contacts and scans an object, tactile vibration is caused due to the texture characteristics of the fine protrusions on the surface of the object, namely, the shearing force and the pressure detected by the upper bionic skin layer 8 are dynamically changed, so that the output light intensity is changed along with the dynamic change. The identification of the texture features can be realized by calibrating the mapping relation between the change characteristics of the output light intensity and the texture features.

Although embodiments of the present utility model have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made therein without departing from the principles and spirit of the utility model, the scope of which is defined in the appended claims and their equivalents.

Claims

1. A texture recognition sensing device based on optical fiber bionic skin, comprising: light source (1), optical fiber bionic skin (2), photoelectric detector (3), digital acquisition card (4), computer (5) and signal processing device (6), its characterized in that: the optical fiber bionic skin (2) is characterized in that two ends of the optical fiber bionic skin (2) are respectively connected with a light source (1) and a photoelectric detector (3), the photoelectric detector (3) is connected with a digital acquisition card (4), the digital acquisition card (4) is connected with a computer (5), and the computer (5) is connected with a signal processing device (6).

2. The texture recognition sensing device based on optical fiber bionic skin according to claim 1, wherein: the light source (1) is used for inputting light to the optical fiber bionic skin (2), and the photoelectric detector (3) is used for receiving and detecting output light.

3. The texture recognition sensing device based on optical fiber bionic skin according to claim 1, wherein: the signal processing means (6) are adapted to identify texture features of the surface of the target object.

4. The texture recognition sensing device based on optical fiber bionic skin according to claim 1, wherein: the optical fiber bionic skin (2) comprises a bionic fingerprint structure (7), an upper bionic skin layer (8), a side bionic skin layer (9) and a substrate (12), wherein the surface of the upper bionic skin layer (8) is provided with a fingerprint-shaped bionic fingerprint structure (7), and the side bionic skin layer (9) is arranged between the upper bionic skin layer (8) and the substrate (12).

5. The texture recognition sensing device based on optical fiber bionic skin according to claim 4, wherein: the side bionic dermis layer (9) comprises a packaging layer (10) and micro-nano optical fiber bionic neurons (11), and the micro-nano optical fiber bionic neurons (11) penetrate through two ends of the packaging layer (10).

6. The texture recognition sensing device based on optical fiber bionic skin according to claim 5, wherein: the micro-nano optical fiber bionic neuron (11) comprises a single-mode fiber region (1101), a micro-nano optical fiber conical transition region (1102) and a micro-nano optical fiber uniform waist region (1103), wherein the two ends of the micro-nano optical fiber uniform waist region (1103) are integrally connected with the micro-nano optical fiber conical transition region (1102), and the other end of the micro-nano optical fiber conical transition region (1102) is integrally connected with the single-mode fiber region (1101).