A device and method for identifying the texture type of kitchen waste

Abstract

The present application aims at the problem that it is difficult to distinguish the texture type of kitchen waste, and provides a device and method for distinguishing the texture type of kitchen waste, which comprises a shearing and crushing device, a diffuse reflection sensor for sensing the feeding of kitchen waste, a Hall sensor on the side wall of the shearing and crushing device, a camera and a microphone array, and a signal processing module for processing the image and sound signal collected by the camera and the microphone array. The effective data formed after the processing of the signal processing module enters the distinguishing program module and classifies the texture type of the kitchen waste through a specific processing flow. The method for distinguishing the texture type of kitchen waste by the interaction of sound signals and image signals can effectively avoid the influence of light environment on the effective data obtained by using the image recognition method. The specific properties of different texture kitchen waste in the processing process are used as the basis for distinguishing, and the specific signal processing process and distinguishing logic can achieve good distinguishing effect.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of kitchen waste treatment, in particular to a device and method for identifying the texture type of kitchen waste. BACKGROUND

[0002] During the pretreatment of kitchen waste, various texture types of kitchen waste need to be dealt with, such as leftover vegetable waste before and after cooking and kitchen waste after eating. In addition, China has a vast territory, and the eating habits are different in different regions, resulting in large differences in the texture of kitchen waste. Since the working parameters of kitchen waste treatment equipment are greatly different when dealing with kitchen waste with large texture differences, the current method mainly classifies at the collection stage, causing great waste of manpower and space.

[0003] In view of the problems of poor accuracy of color identification, great influence of light on image recognition, and great influence of image blocking during feeding, the device and method for identifying the texture type of kitchen waste by collecting sound characteristics through a microphone array and interactive identification with the image processing result can effectively solve the above problems. SUMMARY

[0004] The purpose of the present application is to accurately and automatically identify the texture of kitchen waste, and provide a device and method for identifying the texture type of kitchen waste, which can accurately and automatically identify different textures of post-meal waste and pre-cooking leftover vegetable waste.

[0005] A device and method for identifying the texture type of kitchen waste, the technical scheme adopted includes: a kitchen waste shearing and crushing device, a diffuse reflection sensor installed in the projection area above the shearing and crushing device within a distance of 1 meter from the shearing and crushing device, a Hall sensor installed on the side wall of the shearing and crushing device, the sensing end of the Hall sensor being less than 3mm from the tooth tip of the crushing tooth in the shearing and crushing device, a camera and a microphone array being installed above the shearing and crushing device, a signal processing module connecting the waveform signal, frequency signal and loudness signal in the sound characteristics collected by the microphone array, the signal processing module connecting the image signal collected by the camera, the signal processing module connecting the high and low levels collected by the Hall sensor, and a discrimination program module connecting the signal processing module.

[0006] The signal processing module includes: a binary single-chip microcomputer for binary processing of the image signal collected by the camera, a sound signal collector for separating the sound signal collected by the microphone array into a waveform signal, a frequency signal and a loudness signal, a Hall sensor for sending high and low levels when sensing or not sensing the crushing tooth, and a sound source space-time positioning for judging the time and space of the kitchen waste crushing sound through the loudness signal and the high and low levels to ensure that the collected audio is the kitchen waste crushing sound.

[0007] The method for judging the time and space of the kitchen waste crushing sound by the sound source time and space positioning is that the loudness of the sound collected by the microphones at different positions in the microphone array is different, and the sound signal collected is determined to be emitted by the kitchen waste crushing position through the loudness signal, and the tooth tip of the crushing tooth in the shearing and crushing device starts to crush the kitchen waste, and the Hall sensor senses the high level emitted by the crushing tooth, and the sound source time and space positioning takes the start / end time of the effective sound source signal as the start / end time.

[0008] The discrimination program module includes the fractal dimension value obtained after the binary image data is processed by fractal processing, and the interactive discrimination process uses the fractal dimension value, the waveform signal, the frequency signal and the loudness signal as the data source to discriminate the texture type of the kitchen waste by using specific logic and process.

[0009] The discrimination method of the device for discriminating the texture type of the kitchen waste includes the specific logic and process used by the discrimination program module of the device.

[0010] (1) The sound signal processed by the sound source time and space positioning includes not only the sound emitted by the kitchen waste being crushed but also the friction and collision sound emitted by the intermeshing crushing teeth at this position, the frequency characteristics of the sound signal are judged, the invalid frequency band is discarded, and the frequency band of the sound signal emitted by the kitchen waste crushing is retained to obtain effective audio data.

[0011] (2) According to the characteristics that the post-dinner waste in the kitchen waste is generally small in particle size but the vegetable waste is large in outline, the fractal dimension value obtained by fractal analysis of the kitchen waste has the characteristics that the fractal dimension of the vegetable waste is small and the fractal dimension of the post-dinner waste is large.

[0012] (3) According to the characteristics that the sound loudness of brittle and hard objects being sheared is large and the outline is simple with small image fractal dimension, the loudness and fractal dimension are used as evaluation criteria to judge the effective audio data and the fractal dimension value of the image, and if the loudness is greater than the set value and the fractal dimension is less than the set value, it is determined to be bone or shell waste with brittle and hard texture.

[0013] (4) According to the characteristics that the outline of plant stems and leaves is simple, the cutting performance is good, and there is no adhesion sound, the waveform and fractal dimension are used as evaluation criteria to judge the effective audio data and the fractal dimension value of the image, and if the fractal dimension and the waveform signal are both less than the set value, it is determined to be fruit and vegetable waste with large particles and no viscosity.

[0014] (5) According to the characteristics that the post-dinner waste is small in particle size and has large viscosity, the fractal dimension value is large and there is obvious adhesion sound when it is sheared, the waveform and fractal dimension are used as evaluation criteria to judge the effective audio data and the fractal dimension value of the image, and if the fractal dimension and the waveform signal are both greater than the set value, it is determined to be post-dinner waste with small particles and viscosity.

[0015] Compared with the prior art, the present application has the following advantages:

[0016] Firstly, in the process of identifying the texture type of kitchen waste, the present application adopts an image recognition method to effectively solve the problems of low recognition accuracy caused by poor light environment, device aging caused by long-time operation of the image acquisition system, and large energy consumption of the control system. The device is awakened by collecting the diffuse reflection signal and sound signal during the crushing of kitchen waste, effectively solving the problems of energy consumption and device aging. The sound signal is used as the time reference for the operation of the device. The sound signal is collected during the crushing of kitchen waste, and the sound characteristics are extracted and interacted with the image signal to judge the texture type of kitchen waste, which can effectively avoid the influence of light environment on the image recognition method.

[0017] Secondly, the method adopted by the present application distinguishes the sound characteristics of kitchen waste with different textures during the crushing process due to viscosity, rheological properties, and mechanical properties, and uses the unique properties as the basis for identification. Combined with specific signal processing process and discrimination logic, good discrimination effect can be achieved. BRIEF DESCRIPTION OF DRAWINGS

[0018] Figure 1 is a schematic diagram of the structure of the device of the present application.

[0019] Figure 2 is a logic and flowchart used by the discrimination program module. EMBODIMENT

[0020] As Figure 1 shown, a device for identifying the texture type of kitchen waste includes a kitchen waste shearing and crushing device 01, a diffuse reflection sensor 02 installed in the projection area above the shearing and crushing device within a distance of 1 meter from the shearing and crushing device 01, a Hall sensor 03 installed on the side wall of the shearing and crushing device 01, the sensing end of the Hall sensor 03 being less than 3 mm from the tooth tip of the crushing tooth 06 in the shearing and crushing device 01, a camera 04 and a microphone array 05 being installed above the shearing and crushing device 01, a signal processing module 10 being used to process the image and sound signal collected by the camera 04 and the microphone array 05, the effective data formed after processing by the signal processing module 10 entering a discrimination program module 20, and the texture type of kitchen waste being classified through a specific image signal and sound signal processing process.

[0021] The signal processing module 10 includes: a binarization microcontroller 11 that performs binarization processing on the image signal acquired by the camera 04. The binarized image reduces the storage size and preserves the outline of the kitchen waste, providing input conditions for subsequent fractal processing; an acoustic signal acquisition unit 12 that separates the acoustic signal acquired by the microphone array 05 into waveform signal 15, frequency signal 16, and loudness signal 17; a Hall sensor 03 that emits high and low level signals 13 when the crushing tooth 06 is detected; and a sound source spatiotemporal positioning unit 14 that uses the loudness signal 17 and the high and low level signals 13 to make spatiotemporal judgments on the sound of kitchen waste crushing to ensure that the collected audio is emitted by the crushing of kitchen waste.

[0022] Because the microphone array can collect not only the sound of food waste crushing but also the sound emitted by the mechanical and electrical components of the device itself when the food waste treatment device is working, it is necessary to filter out unnecessary noise from the sound source. The method of spatiotemporal positioning of the sound source 14 for spatiotemporal judgment of the sound of food waste crushing is as follows: The spatial location of the sound is determined by the fact that the loudness and arrival time of the sound emitted from a certain location are different when the microphones at different positions in the microphone array collect the sound. The difference between the time and the loudness signal 17 determines whether the collected sound signal is emitted from the food waste crushing location. If it is not emitted from the food waste crushing location, it can be judged as environmental noise and the sound signal is removed. The time of the sound is determined by the fact that when the tip of the crushing tooth 06 in the shearing crushing device 01 begins to crush the food waste, the Hall sensor 03 senses that the crushing tooth 06 emits a high level. The spatiotemporal positioning of the sound source 14 uses this as the start / end time of the effective sound source signal for calibration. Sound emitted outside this time period can be judged as environmental noise.

[0023] The identification module 20 includes: after binarization, the image data is processed by fractal processing 21 to obtain fractal dimension value 224. Under good lighting conditions, the fractal dimension of the food waste and the leftover food waste can be used to identify them. When the lighting conditions are poor, the interactive identification process 22 is required to use the fractal dimension value 224, waveform signal 15, frequency signal 16, and loudness signal 17 as data sources and use specific logic and process to identify the texture type of the food waste.

[0024] like Figure 2 As shown, the specific logic and process used by the identification program module 20 include:

[0025] (1) The sound signal after being processed by the sound source spatiotemporal positioning 14 includes not only the sound emitted when the kitchen waste is crushed, but also the friction and collision sound emitted by the meshing crushing teeth 06 at that position. The frequency characteristics of the sound signal are judged, and the invalid frequency segment is discarded and the frequency segment of the sound signal emitted by the crushing of kitchen waste is retained to obtain effective audio data.

[0026] (2) According to the characteristics that the post-dinner garbage in the kitchen garbage has smaller particles but the outer shape of the vegetable garbage is larger, the fractal dimension value 224 of the kitchen garbage obtained by fractal analysis has the characteristics that the fractal dimension of the vegetable garbage is smaller and the fractal dimension of the post-dinner garbage is larger.

[0027] (3) According to the characteristics that the brittle and hard object has larger sound loudness and simpler image fractal dimension when being cut, the sound loudness and the fractal dimension are used as evaluation criteria to judge the fractal dimension value of the effective audio data and the image, and the brittle and hard object such as bone or shell can be determined when the sound loudness is larger than a set value and the fractal dimension is smaller than a set value.

[0028] (4) According to the characteristics that the plant stem and leaf have simpler outline, better cutting performance and no adhesion sound, the waveform and the fractal dimension are used as evaluation criteria to judge the fractal dimension value of the effective audio data and the image, and the fruit and vegetable garbage with larger particles and no viscosity can be determined when the fractal dimension and the waveform signal are both smaller than a set value.

[0029] (5) According to the characteristics that the post-dinner garbage has smaller particles and larger viscosity, the fractal dimension value is larger and the adhesion sound is obvious when being cut, the waveform and the fractal dimension are used as evaluation criteria to judge the fractal dimension value of the effective audio data and the image, and the post-dinner garbage with smaller particles and viscosity can be determined when the fractal dimension and the waveform signal are both larger than a set value.

[0030] Since the kitchen garbage itself has no fixed pattern and no fixed sound when being broken, the above embodiments are only used to illustrate the present application, and any equivalent transformation and improvement based on the technical scheme of the present application should not be excluded from the protection scope of the present application.

Claims

1. An apparatus for discriminating a texture type of kitchen waste, characterized by, The kitchen waste shearing and crushing device, a diffuse reflection sensor installed in the projection area above the shearing and crushing device within a range of 1 meter from the shearing and crushing device, a Hall sensor installed on the side wall of the shearing and crushing device, the sensing end of the Hall sensor being less than 3 mm away from the tooth tip of the crushing tooth in the shearing and crushing device, a camera and a microphone array being installed above the shearing and crushing device, a signal processing module connecting the microphone array to collect waveform signals, frequency signals and loudness signals in the sound characteristics, the signal processing module connecting the camera to collect image signals, the signal processing module connecting the Hall sensor to collect high and low levels, and a discrimination program module connecting the signal processing module; The signal processing module comprises: The binary single-chip performs binary processing on the image signals collected by the camera, the sound signal collector separates the sound signals collected by the microphone array into waveform signals, frequency signals and loudness signals, the Hall sensor sends high and low levels when it senses or does not sense the crushing tooth, and the sound source space positioning judges the time and space of the kitchen waste crushing sound according to the loudness signals and the high and low levels, so as to ensure that the collected audio is the kitchen waste crushing sound; The method for judging the time and space of the kitchen waste crushing sound comprises: The loudness of the sound collected by the microphones at different positions in the microphone array is different, and the sound signal collected is determined to be emitted by the kitchen waste crushing position according to the loudness signal, when the tooth tip of the crushing tooth in the shearing and crushing device starts to crush the kitchen waste, the Hall sensor senses the crushing tooth and sends a high level, and the sound source space positioning takes the start / end time of the effective sound source signal as the criterion for suppression; The discrimination program module comprises: The fractal dimension value is obtained after the binary image data is subjected to fractal processing, the fractal dimension value, the waveform signal, the frequency signal and the loudness signal are used as data sources to identify the texture type of the kitchen waste by using specific logic and processes in the interactive discrimination process. The specific logic and processes used by the discrimination program module comprise:

2. The discrimination method of the apparatus for discriminating a texture type of the kitchen garbage according to claim 1, characterized in that, (1) The sound signal processed by the sound source space positioning not only includes the sound emitted by the kitchen waste being crushed, but also includes the friction and collision sound emitted by the intermeshing crushing teeth at the position, the frequency characteristics of the sound signal are judged, the invalid frequency band is discarded, and the frequency band of the sound signal emitted by the kitchen waste crushing is retained to obtain effective audio data; (2) According to the characteristics that the post-meal waste in the kitchen waste is generally small in particle size but the vegetable waste is large in outline, the fractal dimension value obtained by fractal analysis of the kitchen waste has the characteristics that the fractal dimension of the vegetable waste is small and the fractal dimension of the post-meal waste is large; (3) According to the characteristics that the sound emitted by brittle and hard objects being sheared is loud and the outline is simple, the fractal dimension of the image is small, the loudness and the fractal dimension are used as evaluation criteria to judge the effective audio data and the fractal dimension value of the image, and if the loudness is greater than a set value and the fractal dimension is less than a set value, it is determined that the object is bone or shell waste with brittle and hard texture; (4) According to the characteristics that the outline of plant stems and leaves is simple, the cutting performance is good, and there is no adhesion sound, the fractal dimension and the waveform are used as evaluation criteria to judge the effective audio data and the fractal dimension value of the image, and if both the fractal dimension and the waveform signal are less than a set value, it is determined that the object is fruit and vegetable waste with large particles and no viscosity. ​ (5) According to the characteristics of the post-dinner garbage particles being small and having great viscosity, the fractal dimension value is large and the particles have obvious adhesion sound when being sheared. The fractal dimension value of the effective audio data and image is judged according to the evaluation criteria of the waveform and fractal dimension. If the fractal dimension and the waveform signal are both greater than the set value, it can be determined that the post-dinner garbage particles are small and have viscosity.

Images