WO2020087717A1 - 一种基于机器视觉的动物机器人刺激参数测定系统和方法 - Google Patents
一种基于机器视觉的动物机器人刺激参数测定系统和方法 Download PDFInfo
- Publication number
- WO2020087717A1 WO2020087717A1 PCT/CN2018/123659 CN2018123659W WO2020087717A1 WO 2020087717 A1 WO2020087717 A1 WO 2020087717A1 CN 2018123659 W CN2018123659 W CN 2018123659W WO 2020087717 A1 WO2020087717 A1 WO 2020087717A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- stimulation
- measurement
- animal
- robot
- animal robot
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Ceased
Links
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J19/00—Accessories fitted to manipulators, e.g. for monitoring, for viewing; Safety devices combined with or specially adapted for use in connection with manipulators
- B25J19/0095—Means or methods for testing manipulators
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
- G06T3/4038—Image mosaicing, e.g. composing plane images from plane sub-images
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30241—Trajectory
Definitions
- the present invention belongs to the field of animal robot control, and specifically relates to a machine vision-based animal robot stimulation parameter measurement system and method.
- Animal robots refer to the use of animal motility to stimulate the sensory afferent nerves of animals to achieve artificial control or guide certain behaviors of animals, also known as "cyborg".
- Animal robots use the brain Computer Interface (Brain Computer Interface, BCl) technology, which realizes direct information interaction between external control and biological brain regions to complete the precise control of animal robot behavior.
- BCl Brain Computer Interface
- animal robots use biology as their The driving carrier greatly simplifies the motion control and other modules in robot design.
- animal robots can rely on their biological instincts to quickly and effectively deal with emergencies when facing emergencies or complex geographic environments. The flexibility, environmental adaptability and concealment. And animal robots rely on their foraging behavior and do not need external equipment to provide the energy required for sports, which greatly reduces the robot's energy consumption and improves battery life.
- the so-called animal robot is a smart animal that takes a living animal as its body and controls its brain nerves or muscles with encoded electrical signals to control the body.
- the electrical stimulation parameters required to control animal robots are different. Therefore, it is an important part in the research process of animal robots to obtain experimental data for each stimulation channel of each animal robot through the stimulation experiment, and we call this experiment part "stimulation parameter measurement experiment”. This experiment is very important for animal robot research. It will provide a priori information for the later practical research of animal robots.
- the experimenter needs to continuously adjust the stimulation parameters and continuously observe the animal's movement behavior in accordance with his own operation experience, and at the same time record the stimulation parameter data during the experiment to determine the most suitable Stimulation parameters.
- this working method has many shortcomings: First, the experiment has randomness that varies from person to person, and different experimenters have due to their experience and understanding, etc. Differences in aspects, coupled with the subjectivity of the experimenters, make it impossible to obtain consistent recording results in the same experiment process, such as recording the rotation angle of the animal robot under electrical stimulation with specific parameters, and judging the rotation angle by the human eye, deviations are inevitable.
- the present invention proposes a machine vision-based animal robot stimulation parameter measurement system and method, which has a reasonable design, overcomes the deficiencies of the prior art, and has good effects.
- An animal robot stimulation parameter measurement system based on machine vision includes a measurement and control system and a stimulator; the measurement and control system is mainly composed of an industrial camera, a PC and a wireless communication module A; the stimulator is mainly composed of a microprocessor and multi-channel coding Composed of signal generator, wireless communication module B and basic functional circuit;
- the industrial camera is connected to the PC through the USB interface, and is configured to collect the motion state of the animal robot during the stimulation experiment, and transmit the video data to the PC;
- the wireless communication module A is connected to the PC through a serial port, and is configured to wirelessly transmit data from the PC;
- the PC uses a data analysis system developed based on the OpenCV library to analyze and process the transmitted stimulus signals and detection screens, and analyzes the state by calculating the characteristic points of the rat robot to calculate the rat robot The angle of rotation; combined with the stimulation signal corresponding to each rotation, screen out the video frames that effectively control the rat robot, and automatically save these controllable videos; finally, the stimulation signal Analyze and compare the intensity and rotation angle to obtain the corresponding relationship between the rotation angle of the tested rat robot and the stimulation intensity, and calculate the controllable sensitivity;
- the wireless communication module B connected to the microprocessor through a serial port, is configured to wirelessly receive data from a PC;
- the microprocessor is connected to the wireless communication module B through a serial port, and at the same time, controls the working state of the multi-channel coded signal generator through the I / O port; the microprocessor generates the desired coded electricity based on the information received by the wireless communication module B Stimulate the signal and apply it to the target brain area of the animal robot to make the animal produce the desired running behavior;
- the multi-channel coded signal generator is configured to process and transform the original signal of the microprocessor, and realize the selection, start and stop functions of the stimulation channel under the control of the microprocessor, thereby generating a stimulus Coded stimulation signal with adjustable signal frequency, stimulation signal amplitude and stimulation signal time course, and apply this coded stimulation signal to the target brain area;
- the basic functional circuit is configured to provide power to the microprocessor and the wireless communication module B;
- the PC-based measurement and control system generates stimulation parameters and control commands according to the set rules and sends them to the stimulator through the wireless communication device A; meanwhile, the PC-based measurement and control system simultaneously records and saves the stimulation parameters during the experiment Information and video files reflecting the controlled behaviors of animal robots; Finally, based on the video files and corresponding stimulation parameters, the correspondence between the controlled behaviors and the stimulation parameters is analyzed.
- an identification block is provided on the stimulator, the identification block is configured to standardize the measurement of the steering angle of the animal robot during the stimulation experiment, in the actual steering angle measurement, the average of the steering angles of the three identification lines is taken The value is used as the final steering angle; the identification block is composed of three lines of different colors, each line is marked with an arrow used to identify the direction of the line, the three lines intersect at a point, and they are evenly distributed on the plane with a difference of 120 degree.
- both the wireless communication module A and the wireless communication module B select the wireless communication chip NRF9E5; the industrial camera selects a 5 million pixel industrial camera with a USB3.0 data interface; the microprocessor of the stimulator selects the C8 051F410 chip; encoding
- the signal generator consists of 4 sets of symmetrical transistors and 2 MAX309 chips.
- the present invention also refers to a machine vision-based animal robot stimulation parameter measurement method, which uses the machine vision-based animal robot stimulation parameter measurement system as described above, in accordance with the following steps: [0019] Step 1: Install the stimulator with the identification block on the back of the animal robot, its output end is connected to the pre-implanted electrode interface slot, and start the measurement and control system;
- Step 2 Place the animal robot on a plane with a solid background to locate and segment the characteristic lines in the identification block;
- Step 3 Set the parameters; where, the initial parameters are set as follows: the current amplitude is 50uA, the pulse width is 2, the pulse number is 5, the pulse frequency is 90Hz; the maximum parameter settings are as follows: current The amplitude is 130 uA, the pulse width is 9, the pulse number is 30, and the pulse frequency is 130 Hz;
- Step 4 After the initialization of the measurement and control system is completed, firstly, the current position of the animal robot is analyzed and judged according to the real-time image information from the industrial camera, and the type of stimulation is selected; the parameter change of each type of stimulation follows the gradual change of the stimulation intensity from weak to strong The law of increasing;
- Step 5 After the stimulus type is determined, the measurement and control system, according to the progress of the stimulus type, follows the law that the stimulus intensity gradually increases, and sends the parameter data and the stimulus type command to the stimulator together through the wireless communication module A;
- Step 6 After the stimulator is powered on, the microprocessor completes the initialization and waits. After receiving the parameter data and stimulation commands from the measurement and control system through the wireless communication module B, the microprocessor according to the received information, Control the working state of the multi-channel coded signal generator to generate corresponding coded stimulation electrical signals, and apply the stimulation electrical signals to the corresponding brain nerve nucleus of the animal robot to control the movement behavior of the animal robot; It is repeated 10 times within a second; at the same time when the stimulus is generated, the stimulator sends the stimulus start indicator to the measurement and control system through the wireless communication module B.
- the measurement and control system After receiving the stimulus start indicator, the measurement and control system starts the video recording function to start video recording, and saves the video files by category In three different folders; after the end of the stimulation, the stimulator sends the end identification signal to the measurement and control system, and then enters the waiting state. After the measurement and control system receives the end identification signal, the video recording ends after a delay of 2 seconds;
- Step 7 After the end of all experiments, analyze the video data through a PC, draw the animal robot trajectory according to the video file, the measurement and control system reads each frame of the digital image in the video file, and locates according to the specific color in the identification block Draw the intersection point of the three lines, and calculate the coordinates of the intersection point in the image, and finally connect the coordinates of the intersection point in each frame to get the trajectory of the animal robot; at the same time, take the first frame and the video file from the video file. The last frame of the digital image is processed for the first frame of the image, three identification lines in the image are extracted according to the color information, and the direction of each line on the image plane is calculated The angles are respectively denoted as A, B, and C; the last frame of the image is processed in the same way, and the direction angles are respectively denoted as a, b, and c; then the steering angle of the animal robot is ((aA) + (BB) + (cC)) / 3;
- Step 8 After completing the above processing, use image processing technology to stitch together the trajectory images of the same stimulus with the same stimulus parameters, and display the steering angle on each corresponding trajectory map to form an image, according to the image The consistency of the control results and the stability of the control effect under the same stimulation parameters are tested, and finally the results with stability are selected as the parameters.
- the type of stimulation is divided into three types: left-handed stimulation type, right-handed stimulation type and forward stimulation type; the measurement and control system according to the position of the animal during the experiment and each type of experiment Select the stimulation type for the process: If the right side of the animal robot is close to the edge of the field, select the left-turn stimulation type; if the left side of the animal robot is close to the edge of the field, then select the right-turn stimulation type; if the animal robot is not at the edge of the field, the measurement and control system will Turn right and advance the experiment progress of each of the three stimulus types. Select the stimulus type with the slower progress for the forward stimulus experiment until all three stimulus types are completed.
- the law of gradual increase in stimulation intensity is: in order of amplitude, width, number, and frequency, each parameter change alternately increases one of the above four parameters, each The increments of the variables are as follows: the current amplitude increment is 5uA, the pulse width increment is 1, the pulse number increment is 5, the pulse frequency increment is 10Hz; until the four variables related to the stimulation intensity reach the set The maximum value, when the parameter of a certain stimulation type reaches the maximum value, it indicates that this type of experiment has completed one cycle test, and then the next cycle test is performed.
- step 6 three different folders are named in the format of stimulation type code + number of cycles + stimulation parameters; the stimulation type codes are: left turn code is 1, right turn code is 2, forward The code is 3; the number of cycles starts from 1, and 1 is added after each cycle is completed; the stimulation parameters are combined in the order of amplitude, pulse width, number of pulses, and stimulation signal frequency.
- the system of the present invention obtains the behavior characteristics of animal robots through machine vision technology, which can make more accurate and quantitative analysis and judgment on the movements of animal robots, and automatically analyze each
- the appropriate stimulation parameter range of the stimulation channel effectively overcomes the previous shortcomings, which not only ensures the standardity of the experimental procedure and the objectivity of the experimental data, obtains objective and consistent stimulation parameter measurement results, but also greatly reduces the stimulation parameter measurement experiment.
- the low-value labor has improved the efficiency of animal robot research.
- FIG. 1 is a schematic structural diagram of an animal robot experimental data collection system based on machine vision.
- FIG. 2 is a schematic diagram of the working principle of an animal robot experimental data collection system based on machine vision.
- FIG. 3 is a flow chart of the measurement and control system.
- FIG. 4 is a flow chart of the stimulator.
- 5 is a schematic diagram of identification line extraction and angle calculation.
- FIG. 6 is a schematic diagram of the experimental results of stimulus parameter determination based on machine vision.
- An animal robot stimulation parameter measurement system based on machine vision includes a measurement and control system and a stimulator; the measurement and control system is mainly composed of an industrial camera, a PC and a wireless communication module A; the stimulator is mainly composed of a micro Composed of a processor, a multi-channel coded signal generator, a wireless communication module B, and basic functional circuits; [0040] An industrial camera, connected to a PC via a USB interface, is configured to collect the motion state of the animal robot during the stimulation experiment , Transfer video data to PC;
- the wireless communication module A connected to the PC through a serial port, is configured to wirelessly transmit data from the PC;
- a PC uses a data analysis system based on the OpenCV library to analyze and process the transmitted stimulus signals and detection screens, and analyzes the state by calculating the characteristic points of the rat robot to calculate the rat robot Rotation angle; combined with the stimulation signal corresponding to each rotation, screen out the video frames that effectively control the rat robot, and automatically save these controllable videos; finally, analyze and compare the stimulation signal intensity and the rotation angle to obtain The corresponding relationship between the rotation angle of the tested rat robot and the stimulus intensity to calculate the controllable sensitivity;
- the wireless communication module B connected to the microprocessor through a serial port, is configured to wirelessly receive data from a PC;
- the microprocessor is connected to the wireless communication module B through a serial port, and at the same time, controls the working state of the multi-channel coded signal generator through the I / O port; the microprocessor generates the desired coded electricity based on the information received by the wireless communication module B Stimulate the signal and apply it to the target brain area of the animal robot to make the animal produce the desired running behavior;
- the multi-channel coded signal generator is configured to process and transform the original signal of the microprocessor, and realize the selection, start and stop functions of the stimulation channel under the control of the microprocessor, thereby generating a stimulus Coded stimulation signal with adjustable signal frequency, stimulation signal amplitude and stimulation signal time course, and apply this coded stimulation signal to the target brain area;
- the basic functional circuit is configured to provide power to the microprocessor and the wireless communication module B;
- the PC-based measurement and control system generates stimulation parameters and control commands according to the set rules, and sends them to the stimulator through the wireless communication device A; meanwhile, the PC-based measurement and control system simultaneously records and saves the stimulation parameters during the experiment Information and video files reflecting the controlled behaviors of animal robots; Finally, based on the video files and corresponding stimulation parameters, the correspondence between the controlled behaviors and the stimulation parameters is analyzed.
- FIG. 2 the working principle of the machine vision-based animal robot experimental data collection system is shown in FIG. 2.
- a sign board with red, green and blue sign lines was installed on the surface of the stimulator and fixed together on the back of the rat.
- the output of the stimulator is connected to the pre-implanted electrode interface slot.
- the industrial camera is fixed directly above the experiment site, and the PC reads the video data from the industrial camera through the USB interface.
- the measurement and control system is responsible for analyzing the image information to determine the position of the rat on the experimental site, and to exchange information with the stimulator through wireless communication to control the working state of the stimulator, while receiving feedback data from the stimulator.
- the workflow of the measurement and control system is shown in FIG. 3.
- the measurement and control system first analyzes the position of the rat in the field based on the real-time image from the industrial camera. If the left side of the rat is near the edge of the field, the measurement and control system selects the right-turn stimulation type ; If the right side of the rat is close to the edge of the field, the measurement and control system selects the left-turn stimulation type; if the rat is located in the center of the field, the measurement and control system selects the stimulation type with a slower progress according to the experimental progress of the three stimulation types.
- the measurement and control system calculates the stimulation parameters required for this test according to the progress of the type, and then passes the parameter data and the stimulation type command together through the wireless communication module A Sent to the stimulator, at the same time, waiting for the stimulator's feedback information, when the start signal is received, the video recording function is started to start recording the robot mouse experimental video, until the end sign received from the stimulator is delayed for 2 seconds, and the video recording will be ended, just The recorded experimental video categories are saved in the specified folder. At the same time, name the video file in the format of “stimulation type code + number of cycles + stimulation parameters”.
- the current stimulation type is left-turn stimulation during the second cycle
- the stimulation parameter is the current amplitude of 100 uA
- the pulse width is 5
- the number of pulses is 20, and the pulse frequency is 90Hz.
- the rest of the information is expanded to 3 digits by adding 0 in the front, that is, the number of cycles 002, current amplitude 100, pulse width 005, pulse number 020, pulse frequency 090.
- the file name of the video is 1002100005020 090. Then enter the next cycle.
- the microprocessor C8051F410 completes the initialization and waits.
- the micro The processor C8051F410 controls the working state of the coded information generator according to the received information, so that it generates corresponding coded stimulation electrical signals, and applies the stimulation signals to the corresponding cerebral nerve nucleus of the rat to control the movement behavior of the rat.
- the above stimulation signal is repeated 10 times within 5 seconds.
- the start mark is sent to the measurement and control system through the wireless communication chip.
- the end mark is sent, and then the standby state is entered, and the above process is repeated.
- the measurement and control system analyzes the recorded video files one by one, the measurement and control system can obtain the stimulation parameters and stimulation type information corresponding to the video file according to the name of the video file, and then read each frame of image data in the video As shown in FIG. 5, mark point 0 is extracted according to the preset color brown, and the coordinates of the point are calculated. Finally, all the point coordinates are connected and displayed on an image to get the movement trajectory of the robot mouse. In addition, read the first frame and the last frame of the video and locate the identification block in the image.
- the three lines in the identification block are extracted according to the color information and calculated
- the direction angle of the blue marking line EF is 180-03.
- the image shown in FIG. 3 is the identification block in the first frame image, and the three angles are denoted as A, B, and C, respectively.
- the direction angles of the three lines in the image identification block of the last frame can be obtained, which are respectively denoted as a, b, and c.
- Animal vision experimental data collection methods and systems based on machine vision using machine vision technology to obtain the movement trajectory and rotation direction of the animal robot, can more accurately and quantitatively analyze and judge the movement behavior of animal robots. Further, the corresponding relationship between the stimulation parameters and the controlled behavior of animal robots is quantitatively analyzed, so as to determine the optimal stimulation parameters suitable for specific animal individuals, and provide necessary information for the practical research of animal robots. Machine vision replaces the human eye. This method not only guarantees the standardity of the experimental procedure and the objectivity of the experimental data, but also obtains objective and consistent measurement results of the stimulation parameters. At the same time, it greatly reduces the original low-value labor brought by the artificial stimulation parameter measurement experiment, and improves the research efficiency of animal robots.
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Robotics (AREA)
- Mechanical Engineering (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Toys (AREA)
- Manipulator (AREA)
Abstract
一种基于机器视觉的动物机器人刺激参数测定系统,属于动物机器人控制领域,该系统利用机器视觉技术获取动物机器人的运动轨迹和转动方向,可更精确和定量化的分析动物机器人的运动行为;并进一步地定量化分析出刺激参数与动物机器人受控行为的对应关系,从而确定出适于特定动物个体的最佳刺激参数,为动物机器人的实用化研究提供必要信息;通过机器视觉代替了人工判断,基于本发明所提出的方法不仅保证了实验流程的标准性和实验数据的客观性,进而得到客观一致的刺激参数测定结果;同时,大大减少了原来由人工刺激参数测定实验所带来的低价值劳动,提高了动物机器人的研究效率。还涉及一种基于机器视觉的动物机器人刺激参数测定方法。
Description
一种基于机器视觉的动物机器人刺激参数测定系统和方法
技术领域
[0001] 本发明属于动物机器人控制领域, 具体涉及一种基于机器视觉的动物机器人刺 激参数测定系统及方法。
背景技术
[0002] 动物机器人是指利用动物的运动机能, 通过对动物的感觉传入神经进行刺激, 实现人为的控制或引导动物的某些行为, 也称“半电子人 (cyborg) 。 动物机器 人利用脑机接口 (Brain Computer Interface, BCl) 技术, 实现外部控制与生物脑 区之间的直接信息交互, 完成精确的控制动物机器人的行为。 相比传统的机械 式机器人, 动物机器人由于利用了生物作为其驱动载体, 大大简化了机器人设 计中的运动控制等模块。 同时动物机器人在面对突发状况或者复杂的地理环境 时, 能够依靠其生物本能对突发状况做出快速有效的处理, 有更高的灵活性、 环境适应性能力和隐蔽性。 而且动物机器人依靠其觅食行为, 不需要外部设备 提供运动所需能源, 极大减少了机器人的能源消耗, 提升续航时间。
[0003] 所谓动物机器人, 就是以活体动物为本体、 用编码电信号控制其大脑神经或肌 肉, 以实现对本体控制的智能动物。 但由于动物个体差别和手术过程中定位误 差, 使得控制动物机器人所需要的电刺激参数具有差异性。 因此, 通过刺激实 验为每一个动物机器人的每一刺激通道获取实验数据进而选定适宜的刺激参数 是动物机器人研究过程中的重要环节, 我们称这一实验环节为“刺激参数测定实 验”。 该实验对动物机器人研究至关重要, 它将为后期的动物机器人实用化研究 提供先验信息。
[0004] 现阶段的刺激参数测定实验, 主要依靠人工操作, 需要实验人员结合自身操作 经验, 不断调整刺激参数并持续观察动物的运动行为, 同时记录实验过程中刺 激参数数据, 进而确定出最合适的刺激参数。 但这种工作方式有很多的不足之 处: 其一, 实验存在因人而异的随意性, 不同的实验人员由于其经验和理解等
方面的差异, 加之实验人员的主观性, 致使在相同实验过程并不能得到一致性 的记录结果, 比如记录特定参数电刺激下的动物机器人转动角度, 通过人眼判 断转动角度, 偏差在所难免。 这种由人为因素带来的不一致性记录结果, 将给 刺激参数与转动角度的对应关系的分析和研究带来不利影响, 最终导致不能得 到两者关系的客观规律。 其二, 由于操作人员注意力、 体力等生理局限无法做 到长时间、 高标准的投入到实验过程中, 时常会漏掉一些关键的测试环节, 尽 管漏掉的测试可以补做, 但已破坏了实验流程的一致性, 没有标准化的实验流 程将不能得到一致的实验结果。 其三, 基于人工操作的刺激参数测定实验消耗 时间长, 同时也增加人力和财力成本。
发明概述
技术问题
问题的解决方案
技术解决方案
[0005] 针对现有技术中存在的上述技术问题, 本发明提出了一种基于机器视觉的动物 机器人刺激参数测定系统及方法, 设计合理, 克服了现有技术的不足, 具有良 好的效果。
[0006] 为了实现上述目的, 本发明采用如下技术方案:
[0007] 一种基于机器视觉的动物机器人刺激参数测定系统, 包括测控系统和刺激器; 测控系统主要由工业相机、 PC机和无线通讯模块 A组成; 刺激器主要由微处理器 、 多通道编码信号发生器、 无线通讯模块 B和基本功能电路组成;
[0008] 工业相机, 通过 USB接口与 PC机连接, 被配置为用于采集刺激实验过程中的动 物机器人的运动状态, 将视频数据传送给 PC机;
[0009] 无线通讯模块 A, 通过串口与 PC机连接, 被配置为用于将来自 PC机的数据无线 发射出去;
[0010] PC机, 使用基于 OpenCV库所开发的数据分析系统, 对传输来的刺激信号和检 测画面进行分析与处理, 通过检测大鼠机器人特征点的方法分析出所处状态, 计算出大鼠机器人转动的角度; 结合每次转动所对应的刺激信号, 筛选出对大 鼠机器人有效控制的录像画面, 自动保存这些可控制的录相; 最后将刺激信号
强度和转动角度进行分析、 比对, 得到被测大鼠机器人的转动角度与刺激强度 之间对应关系, 计算出可控灵敏度;
[0011] 无线通讯模块 B, 通过串口与微处理器连接, 被配置为用于无线接收来自 PC机 的数据;
[0012] 微处理器, 通过串口与无线通讯模块 B连接, 同时, 通过 I/O口控制多通道编码 信号发生器的工作状态; 微处理器基于无线通讯模块 B接收的信息产生期望的编 码电刺激信号, 并将其施加到动物机器人的目标脑区上, 使动物产生期望的运 行行为;
[0013] 多通道编码信号发生器, 被配置为用于对微处理器的原始信号进行处理和变换 , 并在微处理器的控制下实现刺激通道的选择、 启动和停止功能, 进而产生一 个刺激信号频率、 刺激信号幅值和刺激信号时程均可调的编码刺激信号, 并将 这个编码刺激信号施加到目标脑区;
[0014] 基本功能电路, 被配置为用于为微处理器和无线通讯模块 B提供电能;
[0015] 基于 PC机的测控系统按照设定的规律生成刺激参数和控制命令, 并通过无线通 讯设备 A发送给刺激器; 同时, 基于 PC机的测控系统同步记录和保存实验过程中 的刺激参数信息和反映动物机器人的受控行为的视频文件; 最后, 基于视频文 件和相应的刺激参数, 分析出受控行为与刺激参数的对应关系。
[0016] 优选地, 刺激器上设置有标识块, 标识块被配置为用于标准化刺激实验过程中 动物机器人转向角的测量, 在实际的转向角测量中, 取三个标识线转向角的平 均值作为最终的转向角; 标识块由三条不同颜色的线条组成, 每个线条均标有 一个用于标识该线条的方向的箭头, 三个线条交叉于一点, 在平面上平均分布 两两相差 120度。
[0017] 优选地, 无线通讯模块 A和无线通讯模块 B均选择无线通讯芯片 NRF9E5 ; 工业 相机选择具有 USB3.0数据接口的 500万像素工业相机; 刺激器的微处理器选择 C8 051F410芯片; 编码信号发生器由 4组对称的三极管和 2个 MAX309芯片组成。
[0018] 此外, 本发明还提到一种基于机器视觉的动物机器人刺激参数测定方法, 该方 法采用如上所述的基于机器视觉的动物机器人刺激参数测定系统, 按照如下步 骤进行:
[0019] 步骤 1 : 将带有标识块的刺激器安装固定于动物机器人的背上, 其输出端与预 先植入的电极接口插槽连接, 并开启测控系统;
[0020] 步骤 2: 将动物机器人放置于背景为纯色的平面上, 以定位和分割标识块中的 特征线条;
[0021] 步骤 3: 进行参数设定; 其中, 起始参数设定如下: 电流幅值为 50uA, 脉冲宽 度为 2, 脉冲个数为 5, 脉冲频率为 90Hz; 最大值参数设定如下: 电流幅值为 130 uA, 脉冲宽度为 9, 脉冲个数为 30, 脉冲频率为 130Hz;
[0022] 步骤 4: 测控系统初始化完成后, 首先根据来自工业相机的实时图像信息, 分 析判断动物机器人当前的位置, 选择刺激类型; 每种刺激类型的参数变化都遵 循刺激强度由弱到强渐变递增的规律;
[0023] 步骤 5: 刺激类型确定后, 测控系统根据该刺激类型的进度, 遵循刺激强度渐 变递增的规律, 通过无线通讯模块 A将参数数据和刺激类型命令一起发送到刺激 器;
[0024] 步骤 6: 刺激器上电后由微处理器完成初始化后, 进行等待状态, 当通过无线 通讯模块 B接收到来自测控系统的参数数据和刺激命令后, 微处理器根据接收的 信息, 控制多通道编码信号发生器的工作状态, 使其产生相应的编码刺激电信 号, 并将刺激电信号施加于动物机器人对应的脑神经核团, 控制动物机器人的 运动行为; 上述刺激电信号在 5秒内被重复 10次; 产生刺激的同时, 刺激器通过 无线通讯模块 B向测控系统发送刺激开始标识, 测控系统收到刺激开始标识后, 启动录相功能开始录相, 并将视频文件分类保存在三个不同的文件夹中; 本次 刺激结束后, 刺激器发送结束标识信号到测控系统, 然后进入等待状态, 测控 系统收到结束标识信号后, 延迟 2秒后结束录相;
[0025] 步骤 7: 全部实验结束后, 通过 PC机进行视频数据分析, 根据视频文件绘制动 物机器人运动轨迹, 测控系统读取视频文件中的每一帧数字图像, 根据标识块 中的特定颜色定位出三条线的交点, 并计算出该交点在图像中的坐标, 最后将 每一帧中该交点的坐标连接起来, 就可得到动物机器人的运动轨迹; 同时, 从 视频文件中取出第一帧和最后一帧数字图像, 针对第一帧图像进行处理, 根据 色彩信息提取出图像中的三条标识线, 并计算出每一条线在图像平面上的方向
角, 分别记为 A, B, C; 按同样的方式对最后一帧图像进行处理, 方向角分别 记为 a, b, c; 则动物机器人的转向角度为 ( (a-A) + (B-B) + (c-C) ) /3;
[0026] 步骤 8: 完成上述处理后, 利用图像处理技术, 将同类刺激具有相同刺激参数 的轨迹图像, 拼接到一起, 并将转向角显示于每一个对应的轨迹图上形成一个 图像, 根据图像检测相同刺激参数下控制结果的一致性和控制效果的稳定性, 最后将具有稳定性的结果作为选定参数。
[0027] 优选地, 在步骤 4中, 刺激类型分为三种: 左转向刺激类型、 右转向刺激类型 和前进刺激类型; 测控系统根据实验过程中动物所处的位置和每一种类型的实 验进程选取刺激类型: 如果动物机器人的右边靠近场地边缘, 则选择左转向刺 激类型; 如果动物机器人的左边靠近场地边缘, 则选择右转向刺激类型; 如果 动物机器人不在场地边缘, 测控系统根据左转、 右转和前进三种刺激类型各自 的实验进程, 选择进度较慢的那个刺激类型进行前进刺激实验, 直到三种刺激 类型都完成。
[0028] 优选地, 在步骤 4中, 刺激强度渐变递增的规律是: 按幅值、 宽度、 个数和频 率的先后次序, 每次参数变化交替增大上述 4个参数中的一个变量, 各个变量的 增量如下: 电流幅值增量为 5uA, 脉冲宽度增量为 1, 脉冲个数增量为 5 , 脉冲频 率增量为 10Hz; 直到与刺激强度有关的 4个变量都达到设定的最大值, 当某一刺 激类型的参数达到最大值时, 标志着该类实验完成了一个循环测试, 然后再进 行下一个循环测试。
[0029] 优选地, 在步骤 6中, 三个不同的文件夹以刺激类型代码 +循环次数 +刺激参数 的格式命名; 刺激类型代码分别为: 左转代码为 1, 右转代码为 2, 前进代码为 3 ; 循环次数从 1开始, 每完成一次循环后加 1 ; 刺激参数按幅值、 脉冲宽度、 脉 冲个数和刺激信号频率的先后顺序组合而成。
发明的有益效果
有益效果
[0030] 本发明所带来的有益技术效果:
[0031] 本发明系统通过机器视觉技术获取动物机器人的行为特征, 可对动物机器人的 动作做出更加精确和定量化的分析和判断, 并结合本发明方法自动分析出每一
刺激通道的适宜刺激参数范围, 有效克服了先前的不足, 不仅保证了实验流程 标准的性和实验数据的客观性, 得到客观一致的刺激参数测定结果, 而且大大 减轻了由刺激参数测定实验所带来的低价值劳动, 提高了动物机器人的研究效 率。
对附图的简要说明
附图说明
[0032] 图 1为基于机器视觉的动物机器人实验数据采集系统结构示意图。
[0033] 图 2为基于机器视觉的动物机器人实验数据采集系统工作原理示意图。
[0034] 图 3为测控系统工作流程图。
[0035] 图 4为刺激器工作流程图。
[0036] 图 5为标识线提取与角度计算示意图。
[0037] 图 6为基于机器视觉的刺激参数测定实验结果示意图。
发明实施例
本发明的实施方式
[0038] 下面结合附图以及具体实施方式对本发明作进一步详细说明:
[0039] 一种基于机器视觉的动物机器人刺激参数测定系统, 如图 1所示, 包括测控系 统和刺激器; 测控系统主要由工业相机、 PC机和无线通讯模块 A组成; 刺激器主 要由微处理器、 多通道编码信号发生器、 无线通讯模块 B和基本功能电路组成; [0040] 工业相机, 通过 USB接口与 PC机连接, 被配置为用于采集刺激实验过程中的动 物机器人的运动状态, 将视频数据传送给 PC机;
[0041] 无线通讯模块 A, 通过串口与 PC机连接, 被配置为用于将来自 PC机的数据无线 发射出去;
[0042] PC机, 使用基于 OpenCV库所开发的数据分析系统, 对传输来的刺激信号和检 测画面进行分析与处理, 通过检测大鼠机器人特征点的方法分析出所处状态, 计算出大鼠机器人转动的角度; 结合每次转动所对应的刺激信号, 筛选出对大 鼠机器人有效控制的录像画面, 自动保存这些可控制的录相; 最后将刺激信号 强度和转动角度进行分析、 比对, 得到被测大鼠机器人的转动角度与刺激强度 之间对应关系, 计算出可控灵敏度;
[0043] 无线通讯模块 B, 通过串口与微处理器连接, 被配置为用于无线接收来自 PC机 的数据;
[0044] 微处理器, 通过串口与无线通讯模块 B连接, 同时, 通过 I/O口控制多通道编码 信号发生器的工作状态; 微处理器基于无线通讯模块 B接收的信息产生期望的编 码电刺激信号, 并将其施加到动物机器人的目标脑区上, 使动物产生期望的运 行行为;
[0045] 多通道编码信号发生器, 被配置为用于对微处理器的原始信号进行处理和变换 , 并在微处理器的控制下实现刺激通道的选择、 启动和停止功能, 进而产生一 个刺激信号频率、 刺激信号幅值和刺激信号时程均可调的编码刺激信号, 并将 这个编码刺激信号施加到目标脑区;
[0046] 基本功能电路, 被配置为用于为微处理器和无线通讯模块 B提供电能;
[0047] 基于 PC机的测控系统按照设定的规律生成刺激参数和控制命令, 并通过无线通 讯设备 A发送给刺激器; 同时, 基于 PC机的测控系统同步记录和保存实验过程中 的刺激参数信息和反映动物机器人的受控行为的视频文件; 最后, 基于视频文 件和相应的刺激参数, 分析出受控行为与刺激参数的对应关系。
[0048] 以机器人鼠为例说明本发明的实施方式, 基于机器视觉的动物机器人实验数据 采集系统工作原理如图 2所示。 实验时将带有红、 绿、 蓝三色标识线的标识板安 装在刺激器表面, 一起固定于大鼠的背上。 刺激器的输出端与预先植入的电极 接口插槽连接。 工业相机固定于实验场地的正上方, PC机通过 USB接口读取来 自工业相机的视频数据。 测控系统负责分析图像信息, 以判断出大鼠在实验场 地的位置, 并以无线通讯的方式与刺激器交互信息, 控制刺激器的工作状态, 同时接收刺激器的反馈数据。
[0049] 测控系统工作流程如图 3所示, 测控系统首先根据来自工业相机的实时图像分 析大鼠在场地中的位置, 若大鼠的左侧靠近场地的边缘, 测控系统选择右转刺 激类型; 若大鼠的右侧靠近场地的边缘, 测控系统选择左转刺激类型; 若大鼠 位于场地中心测控系统根据三种刺激类型的实验进程, 选择进度较慢的刺激类 型进行刺激实验。 刺激类型确定后, 测控系统根据该类型的进度, 推算出本次 测试所需的刺激参数, 然后通过无线通讯模块 A将参数数据和刺激类型命令一起
发送到刺激器, 同时, 等待刺激器的反馈信息, 当接收到开始信号后, 启动录 相功能开始录制机器人鼠实验视频, 直到接收来自刺激器的结束标识延迟 2秒后 结束录相, 将刚刚录制的实验视频分类保存在指定的文件夹中。 同时, 以“刺激 类型代码 +循环次数 +刺激参数”的格式命名该视频文件, 比如当前的刺激类型为 第二次循环进程的左转刺激, 刺激参数为电流幅值 100 uA, 脉冲宽度为 5 , 脉冲 个数为 20, 脉冲频率为 90Hz。 除刺激类型代码不变外, 其余信息被通过在前方 补 0的方式扩成 3位, 即循环次数 002、 电流幅值 100、 脉冲宽度 005、 脉冲个数为 020、 脉冲频率 090。 将以上数据组合起来, 则该视频的文件名为 1002100005020 090。 然后进入下一次循环。
[0050] 刺激器工作流程如图 4所示, 刺激器上电后由微处理器 C8051F410完成初始化后 , 进行等待状态, 当通过无线通讯模块 B接收到来自测控测控系统的参数和命令 后, 微处理器 C8051F410根据接收的信息控制编码信息发生器的工作状态, 使其 产生相应的编码刺激电信号, 并将刺激信号施加于大鼠对应的脑神经核团, 控 制大鼠的运动行为。 上述刺激信号在 5秒内被重复 10次。 产生刺激的同时, 通过 无线通讯芯片发开始标志到测控系统, 本次刺激结束后发送结束标志, 然后进 入等待状态, 重复以上过程。
[0051] 实验结束后, 测控系统对记录的视频文件进行逐个分析, 测控系统根据视频文 件的名称可以得到与视频文件对应的刺激参数及刺激类型信息, 然后读取视频 中的每一帧图像数据, 如图 5所示根据预设的颜色褐色提取标记点 0点, 并计算 出该点的坐标。 最后将所有的点坐标连接起来, 并显示在一张图像上得到机器 人鼠的运动轨迹。 另外, 读取视频中的第一帧和最后一帧图像并定位分割出图 像中的标识块, 按如图 5所示的方法, 将标识块中的三条线根据颜色信息提取出 来, 并计算出每条线两个端点在图像中的坐标, 以水平右向为参考方向, 根据 线段两端点的坐标可以分别得到红色标识线 CD的角度为 01, 绿色标识线 AB的方 向角为 360-02, 蓝色标识线 EF的方向角为 180-03。 假设图 3所示的是第一帧图像 中的标识块, 三个角度分别记为 A,B,C。 按同样的方法可以得到最后一帧图像标 识块中三条线的方向角, 分别记为 a,b,c。 则机器人鼠的转向角为 an= ( (a-A) + (b-B) + (c-C) ) /3。 若 an大于零则机器人鼠左转了 an度, 若 an小于零则说明
机器人鼠右转了 an度。
[0052] 完成上述处理后, 根据视频文件名称查找出具有相同刺激类型且具有相同刺激 参数的轨迹图像并将其拼接到一起, 并将转向角显示于每一个对应的轨迹图上 形成一个大的图像, 如图 6所示, al a2...nN代表与相应轨迹对应的转向角。 根据 这个图像可以检测相同刺激参数下控制结果的一致性和控制效果的稳定性, 最 后将具有稳定性控制效果的参数筛选出来, 作为被测试机器人鼠的有效刺激参 数保存在电脑文件中, 以方便后期的深入动物机器人实用化研究。
[0053] 基于机器视觉的动物机器人实验数据采集方法和系统, 利用机器视觉技术获取 动物机器人的运动轨迹和转动方向, 可更精确和定量化的分析和判断动物机器 人的运动行为。 进一步地定量化分析出刺激参数与动物机器人受控行为的对应 关系, 从而确定出适于特定动物个体的最佳刺激参数, 为动物机器人的实用化 研究提供必要信息。 机器视觉代替了人眼, 该方法不仅保证了实验流程标准性 和实验数据的客观性, 而且得到客观一致的刺激参数测定结果。 同时, 大大减 轻了原来由人工刺激参数测定实验所带来的低价值劳动, 提高了动物机器人的 研究效率。
[0054] 当然, 上述说明并非是对本发明的限制, 本发明也并不仅限于上述举例, 本技 术领域的技术人员在本发明的实质范围内所做出的变化、 改型、 添加或替换, 也应属于本发明的保护范围。
Claims
权利要求书
[权利要求 i] 一种基于机器视觉的动物机器人刺激参数测定系统, 其特征在于: 包 括测控系统和刺激器; 测控系统主要由工业相机、 PC机和无线通讯 模块 A组成; 刺激器主要由微处理器、 多通道编码信号发生器、 无线 通讯模块 B和基本功能电路组成;
工业相机, 通过 USB接口与 PC机连接, 被配置为用于采集刺激实验 过程中的动物机器人的运动状态, 将视频数据传送给 PC机; 无线通讯模块 A, 通过串口与 PC机连接, 被配置为用于将来自 PC机 的数据无线发射出去;
PC机, 使用基于 OpenCV库所开发的数据分析系统, 对传输来的刺激 信号和检测画面进行分析与处理, 通过检测大鼠机器人特征点的方法 分析出所处状态, 计算出大鼠机器人转动的角度; 结合每次转动所对 应的刺激信号, 筛选出对大鼠机器人有效控制的录像画面, 自动保存 这些可控制的录相; 最后将刺激信号强度和转动角度进行分析、 比对 , 得到被测大鼠机器人的转动角度与刺激强度之间对应关系, 计算出 可控灵敏度;
无线通讯模块 B, 通过串口与微处理器连接, 被配置为用于无线接收 来自 PC机的数据;
微处理器, 通过串口与无线通讯模块 B连接, 同时, 通过 I/O口控制多 通道编码信号发生器的工作状态; 微处理器基于无线通讯模块 B接收 的信息产生期望的编码电刺激信号, 并将其施加到动物机器人的目标 脑区上, 使动物产生期望的运行行为;
多通道编码信号发生器, 被配置为用于对微处理器的原始信号进行处 理和变换, 并在微处理器的控制下实现刺激通道的选择、 启动和停止 功能, 进而产生一个刺激信号频率、 刺激信号幅值和刺激信号时程均 可调的编码刺激信号, 并将这个编码刺激信号施加到目标脑区; 基本功能电路, 被配置为用于为微处理器和无线通讯模块 B提供电能
基于 PC机的测控系统按照设定的规律生成刺激参数和控制命令, 并 通过无线通讯设备 A发送给刺激器; 同时, 基于 PC机的测控系统同步 记录和保存实验过程中的刺激参数信息和反映动物机器人的受控行为 的视频文件; 最后, 基于视频文件和相应的刺激参数, 分析出受控行 为与刺激参数的对应关系。
[权利要求 2] 根据权利要求 1所述的基于机器视觉的动物机器人刺激参数测定系统 , 其特征在于: 刺激器上设置有标识块, 标识块被配置为用于标准化 刺激实验过程中动物机器人转向角的测量, 在实际的转向角测量中, 取三个标识线转向角的平均值作为最终的转向角; 标识块由三条不同 颜色的线条组成, 每个线条均标有一个用于标识该线条的方向的箭头 , 三个线条交叉于一点, 在平面上平均分布两两相差 120度。
[权利要求 3] 根据权利要求 1所述的基于机器视觉的动物机器人刺激参数测定系统 , 其特征在于: 无线通讯模块 A和无线通讯模块 B均选择无线通讯芯 片 NRF9E5 ; 工业相机选择具有 USB3.0数据接口的 500万像素工业相 机; 刺激器的微处理器选择 C8051F410芯片; 编码信号发生器由 4组 对称的三极管和 2个 MAX309芯片组成。
[权利要求 4] 一种基于机器视觉的动物机器人刺激参数测定方法, 其特征在于: 采 用如权利要求 1所述的基于机器视觉的动物机器人刺激参数测定系统 , 按照如下步骤进行:
步骤 1 : 将带有标识块的刺激器安装固定于动物机器人的背上, 其输 出端与预先植入的电极接口插槽连接, 并开启测控系统;
步骤 2: 将动物机器人放置于背景为纯色的平面上, 以定位和分割标 识块中的特征线条;
步骤 3: 进行参数设定; 其中, 起始参数设定如下: 电流幅值为 50uA , 脉冲宽度为 2, 脉冲个数为 5 , 脉冲频率为 90Hz; 最大值参数设定 如下: 电流幅值为 130uA, 脉冲宽度为 9, 脉冲个数为 30, 脉冲频率 为 130Hz;
步骤 4: 测控系统初始化完成后, 首先根据来自工业相机的实时图像
信息, 分析判断动物机器人当前的位置, 选择刺激类型; 每种刺激类 型的参数变化都遵循刺激强度由弱到强渐变递增的规律;
步骤 5: 刺激类型确定后, 测控系统根据该刺激类型的进度, 遵循刺 激强度渐变递增的规律, 通过无线通讯模块 A将参数数据和刺激类型 命令一起发送到刺激器;
步骤 6: 刺激器上电后由微处理器完成初始化后, 进行等待状态, 当 通过无线通讯模块 B接收到来自测控系统的参数数据和刺激命令后, 微处理器根据接收的信息, 控制多通道编码信号发生器的工作状态, 使其产生相应的编码刺激电信号, 并将刺激电信号施加于动物机器人 对应的脑神经核团, 控制动物机器人的运动行为; 上述刺激电信号在 5秒内被重复 10次; 产生刺激的同时, 刺激器通过无线通讯模块 B向 测控系统发送刺激开始标识, 测控系统收到刺激开始标识后, 启动录 相功能开始录相, 并将视频文件分类保存在三个不同的文件夹中; 本 次刺激结束后, 刺激器发送结束标识信号到测控系统, 然后进入等待 状态, 测控系统收到结束标识信号后, 延迟 2秒后结束录相; 步骤 7: 全部实验结束后, 通过 PC机进行视频数据分析, 根据视频文 件绘制动物机器人运动轨迹, 测控系统读取视频文件中的每一帧数字 图像, 根据标识块中的特定颜色定位出三条线的交点, 并计算出该交 点在图像中的坐标, 最后将每一帧中该交点的坐标连接起来, 就可得 到动物机器人的运动轨迹; 同时, 从视频文件中取出第一帧和最后一 帧数字图像, 针对第一帧图像进行处理, 根据色彩信息提取出图像中 的三条标识线, 并计算出每一条线在图像平面上的方向角, 分别记为 A, B , C; 按同样的方式对最后一帧图像进行处理, 方向角分别记为 a, b , c; 则动物机器人的转向角度为 ( (a-A) + (B-B) + (c-C) ) /3 ;
步骤 8: 完成上述处理后, 利用图像处理技术, 将同类刺激具有相同 刺激参数的轨迹图像, 拼接到一起, 并将转向角显示于每一个对应的 轨迹图上形成一个新的图像, 根据图像检测相同刺激参数下控制结果
的一致性和控制效果的稳定性, 最后将具有稳定性的结果作为选定参 数。
[权利要求 5] 根据权利要求 4所述的基于机器视觉的动物机器人刺激参数测定方法 , 其特征在于: 在步骤 4中, 刺激类型分为三种: 左转向刺激类型、 右转向刺激类型和前进刺激类型; 测控系统根据实验过程中动物所处 的位置和每一种类型的实验进程选取刺激类型: 如果动物机器人的右 边靠近场地边缘, 则选择左转向刺激类型; 如果动物机器人的左边靠 近场地边缘, 则选择右转向刺激类型; 如果动物机器人不在场地边缘 , 测控系统根据左转、 右转和前进三种刺激类型各自的实验进程, 选 择进度较慢的那个刺激类型进行前进刺激实验, 直到三种刺激类型都 完成。
[权利要求 6] 根据权利要求 4所述的基于机器视觉的动物机器人刺激参数测定方法 , 其特征在于: 在步骤 4中, 刺激强度渐变递增的规律是: 按幅值、 宽度、 个数和频率的先后次序, 每次参数变化交替增大上述 4个参数 中的一个变量, 各个变量的增量如下: 电流幅值增量为 5uA, 脉冲宽 度增量为 1, 脉冲个数增量为 5 , 脉冲频率增量为 10Hz; 直到与刺激 强度有关的 4个变量都达到设定的最大值, 当某一刺激类型的参数达 到最大值时, 标志着该类实验完成了一个循环测试, 然后再进行下一 个循环测试。
[权利要求 7] 根据权利要求 4所述的基于机器视觉的动物机器人刺激参数测定方法 , 其特征在于: 在步骤 6中, 三个不同的文件夹以刺激类型代码 +循环 次数 +刺激参数的格式命名; 刺激类型代码分别为: 左转代码为 1, 右 转代码为 2, 前进代码为 3 ; 循环次数从 1开始, 每完成一次循环后加 1 ; 刺激参数按幅值、 脉冲宽度、 脉冲个数和刺激信号频率的先后顺序 组合而成。
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811283293.3A CN109262656B (zh) | 2018-10-31 | 2018-10-31 | 一种基于机器视觉的动物机器人刺激参数测定系统和方法 |
| CN201811283293.3 | 2018-10-31 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2020087717A1 true WO2020087717A1 (zh) | 2020-05-07 |
Family
ID=65190945
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/CN2018/123659 Ceased WO2020087717A1 (zh) | 2018-10-31 | 2018-12-25 | 一种基于机器视觉的动物机器人刺激参数测定系统和方法 |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN109262656B (zh) |
| WO (1) | WO2020087717A1 (zh) |
Families Citing this family (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109675171A (zh) * | 2019-03-04 | 2019-04-26 | 中国科学院深圳先进技术研究院 | 动物刺激方法、装置、设备和存储介质 |
| CN110275532B (zh) * | 2019-06-21 | 2020-12-15 | 珠海格力智能装备有限公司 | 机器人的控制方法及装置、视觉设备的控制方法及装置 |
| CN113384234B (zh) * | 2021-07-08 | 2022-10-25 | 中山大学 | 动物三维视觉测量装置及方法 |
| CN113627256B (zh) * | 2021-07-09 | 2023-08-18 | 武汉大学 | 基于眨眼同步及双目移动检测的伪造视频检验方法及系统 |
| CN114532242B (zh) * | 2022-02-16 | 2023-02-28 | 深圳市元疆科技有限公司 | 一种小型动物行为研究实验箱 |
| CN115056235B (zh) * | 2022-05-27 | 2023-09-05 | 浙江大学 | 基于多模态融合定位的大鼠搜救机器人及搜救方法 |
| CN115607143A (zh) * | 2022-11-10 | 2023-01-17 | 大连理工大学 | 一种基于无线实时姿态检测的脑机接口行为调控评价方法 |
| CN115738079A (zh) * | 2022-11-25 | 2023-03-07 | 东北大学 | 一种动物行为高速视频采集及脑功能刺激装置 |
| CN117310508B (zh) * | 2023-11-30 | 2024-02-27 | 山东科技大学 | 一种快速准确测量锂电池电变量的方法 |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5857433A (en) * | 1996-07-22 | 1999-01-12 | John C. Files | Animal training and tracking device having global positioning satellite unit |
| US20030199944A1 (en) * | 2002-02-08 | 2003-10-23 | Chapin John K. | Method and apparatus for guiding movement of a freely roaming animal through brain stimulation |
| CN1775323A (zh) * | 2005-09-30 | 2006-05-24 | 东北大学 | 可遥控运动行为的脑神经电刺激装置 |
| CN101023737A (zh) * | 2007-03-23 | 2007-08-29 | 浙江大学 | Bci动物实验系统 |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100455015C (zh) * | 2006-03-09 | 2009-01-21 | 西安交通大学 | 一种集成视觉监控的多模式无线脑机交互装置 |
| US20140236039A1 (en) * | 2011-10-21 | 2014-08-21 | Commissariat A I'energie Atomique Et Aux Energies Alternatives | Method of Calibrating and Operating a Direct Neural Interface System |
| CN104199446B (zh) * | 2014-09-18 | 2017-01-25 | 山东科技大学 | 机器人鸟飞行可控性测评系统及测评方法 |
| CN107351080B (zh) * | 2017-06-16 | 2020-12-01 | 浙江大学 | 一种基于相机单元阵列的混合智能研究系统及控制方法 |
-
2018
- 2018-10-31 CN CN201811283293.3A patent/CN109262656B/zh active Active
- 2018-12-25 WO PCT/CN2018/123659 patent/WO2020087717A1/zh not_active Ceased
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5857433A (en) * | 1996-07-22 | 1999-01-12 | John C. Files | Animal training and tracking device having global positioning satellite unit |
| US20030199944A1 (en) * | 2002-02-08 | 2003-10-23 | Chapin John K. | Method and apparatus for guiding movement of a freely roaming animal through brain stimulation |
| CN1775323A (zh) * | 2005-09-30 | 2006-05-24 | 东北大学 | 可遥控运动行为的脑神经电刺激装置 |
| CN101023737A (zh) * | 2007-03-23 | 2007-08-29 | 浙江大学 | Bci动物实验系统 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109262656A (zh) | 2019-01-25 |
| CN109262656B (zh) | 2019-05-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2020087717A1 (zh) | 一种基于机器视觉的动物机器人刺激参数测定系统和方法 | |
| CN108806371B (zh) | 一种基于驾考科目训练的智能判定方法及系统 | |
| CN105549486B (zh) | 基于机器视觉的烟草打顶抑芽检测控制系统 | |
| CN116030527B (zh) | 一种基于事件相机的高速明瞳眼动检测和追踪方法及系统 | |
| CN106385728A (zh) | 一种led灯串地址检测系统及方法 | |
| CN109187366A (zh) | 偏振光流控芯片癌细胞快速检测装置与方法 | |
| CN101972149A (zh) | 视触觉测试仪及视触觉敏感性测试方法 | |
| CN104605838A (zh) | 一种心率检测的方法及装置 | |
| CN101551934A (zh) | 驾驶员疲劳驾驶的监测装置及监测方法 | |
| CN110063736A (zh) | 基于MOD-Net网络的眼动参数监测的疲劳检测及促醒系统 | |
| CN116530981A (zh) | 一种基于面部识别气血状态分析系统及方法 | |
| CN103198472A (zh) | 一种重型汽车连杆成品质量检测方法及其检测系统 | |
| CN116746904B (zh) | 自动出液系统、方法及皮肤美容仪 | |
| CN109009718A (zh) | 一种基于电阻抗技术结合手势控制轮椅的方法 | |
| CN104199446B (zh) | 机器人鸟飞行可控性测评系统及测评方法 | |
| CN204765560U (zh) | 一种基于手势识别的视力测试装置 | |
| CN105380590B (zh) | 一种具有眼位检测功能的设备及其实现方法 | |
| CN103605358B (zh) | 道路交通信号控制机功能在线仿真测试装置 | |
| CN109657722A (zh) | 基于深度学习算法的舌苔图像识别方法及系统 | |
| CN206674270U (zh) | 一种具有交接功能的舞台灯光跟踪锚定系统 | |
| CN107292257A (zh) | 基于深度学习的身体部位自动识别磁共振扫描方法与装置 | |
| CN205433877U (zh) | 智能化低温冷等离子体医学治疗设备 | |
| TWI594725B (zh) | Portable pupil measuring device and its measuring method | |
| CN222719886U (zh) | 基于人工智能与无人机的番茄病害识别装置 | |
| CN106850783B (zh) | 跨孔ct自动化采集和远程监控方法 |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 18938743 Country of ref document: EP Kind code of ref document: A1 |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 18938743 Country of ref document: EP Kind code of ref document: A1 |