Device and method for classifying glass articles using acoustic analysis

Abstract

A method for classifying glass articles through acoustic analysis by a classification device is proposed. The method includes: receiving, by a processor, sound data of a tapping sound generated by applying a tapping operation to a glass article; determining, by the processor, the type of the glass article by performing tapping sound analysis on the sound data, wherein the type of the glass article includes inorganic glass and organic glass; if the type of the glass article is determined to be inorganic glass, receiving, by the processor, echo data of an echo triggered by applying an ultrasonic echo operation to the glass article; and determining, by the processor, a further type of the glass article by performing echo attenuation analysis on the echo data, wherein the further type of the glass article includes crystal glass, borosilicate glass, and soda-lime glass.

Description

[0001] Cross - Reference to Related Applications

[0002] This application claims the priority of U.S. Provisional Patent Application No. 17 / 177,399, filed on April 21, 2021, and U.S. Non - Provisional Patent Application No. 17 / 688,845, filed on March 7, 2022. The entire disclosures of the above - mentioned applications are incorporated herein by reference. Technical Field

[0003] The present invention generally relates to classification methods and classification devices. In particular, it relates to methods, devices, and systems for classifying glass articles via acoustic analysis. Background Art

[0004] In theory, due to the characteristics of the vitrification process, glass can be recycled 100%. However, in practice, when dealing with a batch of glass where different types are mixed together, glass recycling is ineffective. For example, the European Union recognized this difficulty and adopted in 2012 the Regulation for End - of - Waste Criteria for Glass Cullet, which strongly emphasizes the separation and purification of glass materials.

[0005] In traditional recycling methods, waste glass is usually sorted by shape, which assumes that the two main categories, namely flat glass and container glass, mainly contain soda - lime glass. However, this is not the case in the current market. Glass recycling companies have found that the current glass recycling paradigm results in a mixture of glass types. This adversely affects the melting and forming processes because melting different types of glass together results in only a small amount of material forming a low - quality product. Thus, the mixed glass types need additional manual sorting before the glass can be processed.

[0006] Currently, the traditional glass automatic classification and sorting solutions are optical sensing and machine vision methods. However, due to the lack of spectral features in waste glass and the fact that each piece of glass can include multiple colors, using these optical sensing and machine vision methods for glass classification results in low accuracy and efficiency.

[0007] Therefore, there is a need in the art for new classification methods other than shape, optical, and machine vision methods to reduce the time - consuming and labor - intensive work required for manual sorting and separating glass and to improve classification accuracy and efficiency. Summary of the Invention

[0008] According to one aspect of the present invention, there is provided a computer-implemented method for classifying glass articles via acoustic analysis by a classification device. The method includes: receiving, by a processor of the classification device, sound data of a tapping sound generated by tapping a glass article from a sensor of the classification device; determining, by the processor, the type of the glass article by performing tapping sound analysis on the sound data, wherein the type of the glass article includes inorganic glass and organic glass; if the type of the glass article is determined to be inorganic glass, receiving, by the processor, echo data of an echo generated by applying an ultrasonic echo operation to the glass article from a transceiver of the classification device; and determining, by the processor, a further type of the glass article by performing echo attenuation analysis on the echo data, wherein the further type of the glass article includes crystal glass, borosilicate glass, and soda-lime glass.

[0009] According to another aspect of the present invention, there is provided a computer-implemented method for classifying glass articles via acoustic analysis by a classification device. The method includes: receiving, by a processor of the classification device, sound data of a tapping sound generated by tapping a glass article from a sensor of the classification device; determining, by the processor, the type of the glass article by performing tapping sound analysis on the sound data, wherein the type of the glass article includes inorganic glass and organic glass; if the type of the glass article is determined to be inorganic glass, receiving, by the processor, echo data of an echo generated by applying an ultrasonic echo operation to the glass article from a transceiver of the classification device; and determining, by performing sound velocity analysis on the echo data, a further type of the glass article, wherein the further type of the glass article includes crystal glass, borosilicate glass, and soda-lime glass.

[0010] According to still another aspect of the present invention, there is provided a classification device for classifying glass articles via acoustic analysis, the classification device including one or more processors configured to execute machine instructions for implementing the above method. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Embodiments of the present invention will be described below with reference to the drawings, wherein:

[0012] Figure 1 A block diagram showing a classification device for classifying glass articles via acoustic analysis according to an embodiment of the present invention is shown;

[0013] Figure 2 A flowchart showing the classification of glass articles via acoustic analysis is shown;

[0014] Figure 3 A flowchart showing the tapping sound analysis is shown;

[0015] Figure 4Shows the schematic diagram of the tapping sound analysis;

[0016] Figure 5 Shows the characteristics of different glass types used in the tapping sound analysis;

[0017] Figure 6 Shows the flowchart of the echo attenuation sound analysis;

[0018] Figure 7 Shows the schematic diagram of the echo attenuation sound analysis;

[0019] Figure 8 Shows the schematic diagram of the transceiver design used for the echo attenuation sound analysis;

[0020] Figure 9A Shows the flowchart of classifying glass articles via sound analysis according to another embodiment of the invention;

[0021] Figure 9B Shows the flowchart of the sound velocity sound analysis;

[0022] Figure 10 Shows the schematic diagram of the sound velocity sound analysis; and

[0023] Figure 11 Shows the schematic diagram of the transceiver design priority requirements used in the echo attenuation analysis and the sound velocity analysis. Detailed Description

[0024] In the following description, methods and classification devices for classifying glass articles by sound analysis and the like are presented as preferred embodiments. It will be apparent to those skilled in the art that variations including additions and / or substitutions can be made without departing from the scope and spirit of the present invention. Specific details may be omitted. However, the disclosure enables those skilled in the art to practice the teachings herein without undue experimentation.

[0025] According to various embodiments of the present invention, Figure 1 Shows the classification device 100. The classification device 100 includes a processor 110, a non-transitory memory circuit 120, a sound generating device 130, a transceiver 140, and a sensor 150.

[0026] The sound generating device 130 is configured to generate a tapping sound by applying a tapping operation to the glass article 200. The transceiver 140 is an ultrasonic transducer. Additionally, the transceiver 140 is also coupled to the sound generating device 130 and controlled by the sound generating device 130 to generate ultrasonic waves. That is, the sound generating device 130 can instruct the transceiver 140 to initiate an echo by applying an ultrasonic echo operation to the glass article 200. During the ultrasonic echo operation, the transceiver 140 is in contact with the glass article.

[0027] The sensor 150 is configured to receive a tapping sound. The transceiver 140 is also configured to receive echoes.

[0028] For example, during a tapping operation, the sound generating device 130 controls a hard object (such as a rod or a hammer) to tap the wall of the glass article 200 to generate a tapping sound. The sensor 150 coupled to the processor 110 receives the tapping sound. The sensor 150 sends sound data SD to the processor according to the received tapping sound.

[0029] In addition, during an ultrasonic echo operation, the sound generating device 130 controls the transceiver 140 attachable to the surface of the glass article 200 to generate and output ultrasonic waves to the glass article 200, so as to receive a corresponding plurality of echoes. The transceiver 140 generates echo data according to the received echoes. The transceiver 140 sends the echo data ED to the processor 110.

[0030] The non-transitory memory circuit 120 is configured to store the sound data SD, the echo data ED or other types of data.

[0031] The processor 110 executes the machine instructions 121 to implement a method for classifying glass articles by acoustic analysis.

[0032] Embodiment 1

[0033] Reference Figure 2 , in step S210, the processor 110 receives the sound data of the tapping sound corresponding to the glass article 200 from the sensor 150. Next, in step S220, the processor 110 determines the type of the glass article by performing tapping sound analysis on the sound data, where the type of the glass article includes plexiglass and inorganic glass. Step S220 includes steps S221, S222, S223, S224, S225 and S228. In other words, the "tapping sound analysis" is used to sort organic glass articles and inorganic glass articles.

[0034] In another embodiment, the type of the glass article further includes soda-lime glass. That is to say, step S220 becomes step S220', in which the processor 110 determines the type of the glass article by performing tapping sound analysis on the sound data, where the type of the glass article includes plexiglass, inorganic glass and soda-lime glass. And compared with step S220, step S220' further includes steps S226 and S227. In other words, the "tapping sound analysis" of the present application can further sort out soda-lime glass through steps S226 and S227.

[0035] Specifically, reference Figure 3, at step S221, the processor 110 identifies the intensity information and frequency information of each data element of the sound data. Next, at step S222, the processor 110 identifies, from all the intensities of the data elements of the sound data according to the intensity information, a target data element having the maximum intensity. Next, at step S223, the processor 110 identifies the target frequency of the target data element according to the frequency information.

[0036] For example, referring to Figure 4 , a 2D domain map PL41 can be established according to the intensity information and frequency information of the sound data. An intensity curve is generated from the data elements of the sound data. There are five intensity pitches in the map PL41, and each intensity pitch has a corresponding pitch data element located at the top. The region with the first intensity pitch is the base region. The other region with the remaining segments is the harmonic region.

[0037] In addition, the pitch with the maximum intensity is the dominant pitch, and the pitch data element of the dominant pitch is the target data element. The target frequency is the frequency of the target data element.

[0038] When the dominant pitch is in the base region, it is considered a base dominance. When the dominant pitch is in the harmonic region, it is considered a harmonic dominance.

[0039] Referring to Figure 5 , Table T51 shows different types of glass (soda-lime glass, borosilicate glass, crystal glass, and plexiglass). The dominant pitch frequencies of inorganic glasses (i.e., soda-lime glass, borosilicate glass, and crystal glass) are higher than 2 kHz. The dominant pitch frequency of plexiglass is not higher than 2 kHz. In addition, soda-lime glass and plexiglass are base dominances, while borosilicate glass and crystal glass are harmonic dominances.

[0040] Thus, we can utilize these features to distinguish some types of glass articles through steps S224 and S226.

[0041] Referring to Figure 3 , at step S224, the processor 110 determines whether the target frequency is higher than a predetermined frequency threshold. The predetermined frequency threshold can be set to 2 kHz.

[0042] If the target frequency is not higher than the predetermined frequency threshold, at step S225, the processor 110 determines that the type of the glass article is plexiglass.

[0043] If the target frequency is higher than the predetermined frequency threshold, at step S228, the processor 110 determines that the type of the glass article is inorganic glass.

[0044] In another embodiment corresponding to step S220, the processor 110 performs step S226 to determine whether the glass article is soda-lime glass or another type of inorganic glass. In step S226, the processor 110 determines whether the target data element corresponds to the first intensity pitch among one or more intensity pitches of the data element. In other words, the processor 110 determines whether the sound data of the glass article is fundamental-dominated or harmonic-dominated through step S226.

[0045] If the target data element corresponds to the first intensity pitch among one or more intensity pitches, in step S227 the processor 110 determines that the type of the glass article is soda-lime glass. Otherwise, the processor 110 determines that the type of the glass article is inorganic glass (step S228) (for example, the specific type of inorganic glass of the glass article cannot be distinguished through step S226). In this context, further analysis is implemented to classify the subtype of the inorganic glass.

[0046] Return Figure 2 , the method continues with step S230, where the processor 110 receives echo data corresponding to the echo of the glass article from the transceiver 140. Next, in step S240, the processor 110 determines to determine the further type of the glass article by performing echo attenuation analysis on the echo data, where the further type of the glass article includes crystal glass, borosilicate glass, and soda-lime glass.

[0047] In particular, referring to Figure 6 , in step S241, the processor 110 identifies the intensity information and time information of each data element of the echo data. Next, the processor 110 identifies one or more pitch data elements corresponding to one or more intensity pitches of the echo data according to the intensity information.

[0048] Next, in step S243, the processor 110 fits an intensity function for each intensity of the pitch data element, thereby obtaining the attenuation coefficient of the intensity function corresponding to the echo data.

[0049] For example, referring to Figure 7 , a 2D domain map PL71 can be established according to the intensity information and time information of the echo data. An intensity curve is generated according to the elements of the echo data. In the map PL71, there are four intensity pitches, and each intensity pitch has a corresponding pitch data element at the top.

[0050] Based on these details, a trend curve can be calculated according to the intensity of the pitch data element. The intensity function A(t) is derived by the processor 110 to fit the trend curve corresponding to the pitch data element. The intensity function A(t) is presented based on the following equation (1):

[0051] A(t) = A(0)e-B(t-C) (1)

[0052] Where A(t) represents the sound intensity as a function of time; A(0) represents the intensity of the first pitch; B represents the attenuation coefficient; and C represents the timing of the first pitch, which is randomly set to 0. The attenuation coefficient depends on the material properties and is related to the acoustic impedance. High acoustic impedance results in a low attenuation coefficient.

[0053] After fitting the intensity function, the processor 110 obtains the attenuation coefficient of the glass article. As shown in Table T71, different types of inorganic glass have different ranges of attenuation coefficients (B). The processor 110 further determines the subtype of the inorganic glass article based on the obtained attenuation coefficient (B). The picking ranges (RG1, RG2, and RG3) can be calibrated and applied using the recycling company's own database.

[0054] For example, the first range RG1 corresponding to soda-lime glass is higher than 0.6 MHz and less than or equal to 1.3 MHz; the second range RG2 corresponding to borosilicate glass is higher than 1.3 MHz and less than or equal to 1.8 MHz; and the third range RG3 corresponding to crystal glass is higher than 1.8 MHz and less than or equal to 4 MHz.

[0055] In step S244, the processor 110 determines whether the attenuation coefficient is within the first range RG1, the second range RG2, or the third range RG3.

[0056] If the attenuation coefficient is within the first range RG1, the processor 110 determines that the type of the glass article is soda-lime glass. If the attenuation coefficient is within the second range RG2, the processor 110 determines that the type of the glass article is borosilicate glass. If the attenuation coefficient is within the third range RG3, the processor 110 determines that the type of the glass article is crystal glass.

[0057] Reference Figure 8 , the requirements for the design of the transceiver 140 are proposed for echo attenuation analysis. As presented by the equation F81 (Equation 2 provided below), the echo collection efficiency is determined accordingly. Figure 8

[0058]

[0059]

[0060]

[0060] As shown in Table T81, there are some priority requirements for the design of transceiver 140 for echo attenuation analysis. For example, for echo attenuation analysis, the number of received echoes should be greater than 3. To meet this requirement (i.e., the number of echoes should be greater than 3), the echo collection efficiency of transceiver 140 should be greater than 80%. To achieve a collection efficiency of more than 80%, the diameter of transceiver 140 should be less than or equal to 3.5 mm, and this requirement for the diameter of transceiver 140 is a priority requirement for echo attenuation analysis.

[0061] In some scenarios, it is difficult to measure the attenuation of the echo intensity of thick glass articles (usually > 1 cm). Therefore, Embodiment 2 below provides sound velocity analysis.

[0062] Embodiment 2

[0063] The hardware and classification device of Embodiment 2 are the same as those of classification device 100 in Embodiment 1.

[0064] Reference Figure 9A , steps S910, S920 (S920’), and S930 are similar to steps S210, S220, and S230 in Figure 2 and are not described in detail here.

[0065] In step S940, the processor 110 determines the further type of the glass article by performing sound velocity analysis on the echo data, where the further type of the glass article includes crystal glass, borosilicate glass, and soda-lime glass.

[0066] Specifically, referring to Figure 9B , in step S941, the processor identifies the intensity information and time information of each data element of the echo data. Next, in step S942, the processor 110 identifies two pitch data elements corresponding to two adjacent intensity pitches of the echo data ED according to the intensity information, and then identifies the time difference between the two identified pitch data elements according to the time information. In other words, in sound velocity analysis, only two echoes are needed to complete the analysis.

[0067] In addition, in step S943, the processor 110 identifies the thickness value of the wall of the glass article 200.

[0068] Next, in step S944, the processor 110 calculates the sound velocity corresponding to the echo data according to the wall thickness value and the identified time difference.

[0069] In addition, referring to Figure 10, a 2D map PL101 can be established based on the intensity information and time information of the echo data. An intensity curve is generated according to the data elements of the echo data. In the map PL101, there are four intensity pitches, and each intensity pitch has a corresponding pitch data element at the top. However, as mentioned above, only two adjacent pitches (such as echo 1 and echo 2) are used in the sound velocity analysis. In other words, the size of the required echo data can be reduced to have data elements capable of constructing two intensity pitches.

[0070] The processor 110 identifies the time T1 corresponding to the pitch data element PE1 and the time T2 corresponding to the pitch data element PE2. Then the processor 110 identifies the time difference T according to the times T1 and T2 d .

[0071] The sound velocity is calculated based on the following equation (3).

[0072]

[0073] where h represents the thickness value; and T d represents the time difference.

[0074] After calculating the sound velocity, at step S945, the processor 110 determines whether the calculated sound velocity is within the first velocity range, the second velocity range, or the third velocity range.

[0075] If the calculated sound velocity is within the first velocity range SRG1, at step S946, the processor 110 determines that the type of the glass article is soda-lime glass.

[0076] If the calculated sound velocity is within the second velocity range SRG2, at step S947, the processor 110 determines that the type of the glass article is borosilicate glass.

[0077] If the calculated sound velocity is within the third velocity range SRG3, at step S948, the processor 110 determines that the type of the glass article is crystal glass.

[0078] For example, as shown in Table T101, the sound velocity range SRG1 corresponding to soda-lime glass is higher than or equal to 6000 m / s and lower than or equal to 6400 m / s; the sound velocity range SRG2 corresponding to borosilicate glass is higher than or equal to 5000 m / s and lower than or equal to 5400 m / s; the sound velocity range SRG3 corresponding to crystal glass is higher than or equal to 5600 m / s and lower than or equal to 5800 m / s.

[0079] Reference Figure 11 , some requirements designed for the transceiver 140 are proposed for echo attenuation analysis and sound velocity analysis.

[0080] As shown in Table T101, the required induction frequency of transceiver 140 varies according to the different wall thicknesses of glass articles.

[0081] Generally, for glass articles with a thickness greater than 0.5 mm, an induction frequency greater than 5 MHz is required to achieve sufficient resolution.

[0082] It should be noted that for thick glass articles (usually > 1 cm), flat samples or samples with a large curvature diameter (usually > 10 cm), sound velocity analysis can be achieved and is most useful in the sorting of industrial or architectural glass waste.

[0083] The functional units and methods of the device according to embodiments of the present disclosure can be implemented using a computing device, a computer processor, or an electronic circuit including but not limited to an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), and other programmable logic devices configured or programmed according to the teachings of the present disclosure. Those skilled in the software or electronics field can easily prepare computer instructions or software code for running on a computer device, a computer processor, or a programmable logic device based on the teachings of the present disclosure.

[0084] All or part of the methods according to embodiments can be executed in one or more computing devices including a server computer, a personal computer, a portable computer, and mobile computing devices such as smart phones and tablet computers.

[0085] Embodiments include a computer storage medium having computer instructions or software code stored therein that can be used to program a computer or a microprocessor to implement any step of the present invention. The storage medium can include but is not limited to a floppy disk, an optical disk, a Blu-ray disc, a digital video disc, an optical disc drive, and a magneto-optical disc, a read-only memory, a random access memory, a flash device, or any type of medium or device suitable for storing instructions, code, and / or data.

[0086] Each of the functional units according to the embodiments can also be implemented in a distributed computing environment and / or a cloud computing environment, where all or part of the machine instructions are executed in a distributed manner by one or more processing devices, and the one or more processing devices are interconnected through a communication network such as an intranet, a wide area network (WAN), a local area network (LAN), the Internet, and other forms of data transmission media.

[0087] The above description of the present invention has been presented for purposes of illustration and description. It is not intended to be exhaustive or to limit the invention to the specific forms disclosed. Many variations and modifications will be apparent to those skilled in the art.

[0088] Embodiments are chosen and described in order to best explain the principles of the present invention and its practical applications, so that other technicians in the field can understand the different embodiments of the present invention and the different variations suitable for specific uses.

Claims

1. A computer-implemented method for classifying glass articles via acoustic analysis by a classification device, comprising: Receiving, by a sensor of the classification device, sound data of a tapping sound corresponding to a glass article from the sensor of the classification device; Determining, by a processor, the type of the glass article by performing tapping sound analysis on the sound data, wherein the type of the glass article includes inorganic glass and organic glass; If the type of the glass article is determined to be inorganic glass, then receiving, by the processor, echo data of an echo corresponding to the glass article from a transceiver of the classification device; and Determining, by the processor, a further type of the glass article by performing echo attenuation analysis on the echo data, wherein the further type of the glass article includes crystal glass, borosilicate glass, and soda-lime glass, wherein the step of determining the type of the glass article by performing tapping sound analysis on the sound data includes: Identifying intensity information and frequency information of each data element of the sound data; Identifying, according to the intensity information, a target data element having the maximum intensity among all intensities of the data elements of the sound data; Identifying a target frequency of the target data element according to the frequency information; Determining whether the target frequency is higher than a predetermined frequency threshold; If the target frequency is not higher than the predetermined frequency threshold, determining that the type of the glass article is organic glass; If the target frequency is higher than the predetermined frequency threshold, determining that the type of the glass article is inorganic glass, and wherein the type of the glass article further includes soda-lime glass, and the step of determining the type of the glass article by performing tapping sound analysis on the sound data further includes: If the target frequency is higher than the predetermined frequency threshold, determining whether the target data element corresponds to a first intensity pitch among one or more intensity pitches of the data elements; If the target data element corresponds to the first intensity pitch, determining that the type of the glass article is soda-lime glass; and If the target data element does not correspond to the first intensity pitch, determining that the type of the glass article is an inorganic glass other than soda-lime glass.

2. The computer-implemented method according to claim 1, wherein the predetermined frequency threshold is 2 kHz.

3. The computer-implemented method according to claim 1, wherein the step of determining a further type of the glass article by performing echo attenuation analysis on the echo data includes: Identifying intensity information and time information of each data element of the echo data; Identifying one or more pitch data elements corresponding to one or more intensity pitches of the echo data according to the intensity information; Fitting an intensity function for each intensity of the pitch data element, thereby obtaining an attenuation coefficient of the intensity function corresponding to the echo data; Determining whether the attenuation coefficient is within a first range, a second range, or a third range; If the attenuation coefficient is within the first range, determining that the type of the glass article is soda-lime glass; If the attenuation coefficient is within the second range, determining that the type of the glass article is borosilicate glass; and If the attenuation coefficient is within the third range, determine that the type of the glass article is crystal glass.

4. The computer-implemented method according to claim 3, wherein the intensity function is presented based on the following equation: A(t) = A(0)e -B(t-C) where A(t) represents the sound intensity of the tone as a function of time; A(0) represents the intensity of the first pitch; B represents the attenuation coefficient; and C represents the timing of the first pitch, which is randomly set to 0.

5. The computer-implemented method according to claim 3, wherein the requirements of the transceiver include: The diameter of the transceiver for receiving the echo is less than 3.5 mm.

6. A computer-implemented method for classifying glass articles via acoustic analysis by a classification device, comprising: Receiving, by a processor of the classification device, sound data of a tapping sound corresponding to a glass article from a sensor of the classification device; Determining, by the processor, the type of the glass article by performing tapping sound analysis on the sound data, wherein the type of the glass article includes inorganic glass and organic glass; If the type of the glass article is determined to be inorganic glass, receiving, by the processor, echo data of an echo corresponding to the glass article from a transceiver of the classification device; and Determining a further type of the glass article by performing sound velocity analysis on the echo data, wherein the further type of the glass article includes crystal glass, borosilicate glass, and soda-lime glass, wherein the step of determining the type of the glass article by performing tapping sound analysis on the sound data includes: Identifying intensity information and frequency information of each data element of the sound data; Identifying a target data element having the maximum intensity among all intensities of the data elements of the sound data according to the intensity information; Identifying a target frequency of the target data element according to the frequency information; Determining whether the target frequency is higher than a predetermined frequency threshold; If the target frequency is not higher than the predetermined frequency threshold, determining that the type of the glass article is organic glass; If the target frequency is higher than the predetermined frequency threshold, determining that the type of the glass article is inorganic glass, and wherein the type of the glass article further includes soda-lime glass, and the step of determining the type of the glass article by performing tapping sound analysis on the sound data further includes: If the target frequency is higher than the predetermined frequency threshold, determining whether the target data element corresponds to a first intensity pitch among one or more intensity pitches of the data elements; If the target data element corresponds to the first intensity pitch, determining that the type of the glass article is soda-lime glass; and If the target data element does not correspond to the first intensity pitch, determining that the type of the glass article is an inorganic glass other than soda-lime glass.

7. The computer-implemented method according to claim 6, wherein the predetermined frequency threshold is 2 kHz.

8. The computer-implemented method according to claim 6, wherein the step of determining a further type of the glass article by performing sound velocity analysis on the echo data includes: Identify the intensity information and time information of each data element of the echo data; Identify two pitch data elements corresponding to two adjacent intensity pitches of the echo data according to the intensity information, and identify the time difference between the two identified pitch data elements according to the time information; Identify the thickness value of the wall of the glass article; Calculate the sound speed corresponding to the echo data according to the thickness value and the identified time difference; Determine whether the calculated sound speed is within a first range, a second range, or a third range; If the calculated sound speed is within the first range, determine that the type of the glass article is soda-lime glass; If the calculated sound speed is within the second range, determine that the type of the glass article is borosilicate glass; and If the calculated sound speed is within the third range, determine that the type of the glass article is crystal glass.

9. The computer-implemented method according to claim 8, wherein the sound speed is calculated based on the following equation: where h represents the thickness value; and T d represents the time difference.

10. A classification device for classifying glass articles via acoustic analysis, comprising: A sound generating device configured to generate a tapping sound by applying a tapping operation to a glass article; A transceiver electrically coupled to the sound generating device, wherein the transceiver is configured to generate an echo by applying an ultrasonic echo operation to the glass article and is further configured to receive the echo, wherein the ultrasonic echo operation is indicated by the sound generating device; A sensor configured to receive the tapping sound; and A processor electrically coupled to the sound generating device, the transceiver, and the sensor, wherein the processor is configured to execute machine instructions to implement a computer-implemented method, the method comprising: Receiving, by the processor, sound data of the tapping sound from the sensor; Determining, by the processor, the type of the glass article by performing tapping sound analysis on the sound data, wherein the type of the glass article includes inorganic glass and organic glass; If the type of the glass article is determined to be inorganic glass, receiving, by the processor, echo data of the echo from the transceiver; and Determining, by the processor, a further type of the glass article by performing echo attenuation analysis on the echo data, wherein the further type of the glass article includes crystal glass, borosilicate glass, and soda-lime glass, wherein the step of determining the type of the glass article by performing tapping sound analysis on the sound data comprises: Identifying the intensity information and frequency information of each data element of the sound data; Identifying a target data element having the maximum intensity among all the intensities of the data elements of the sound data according to the intensity information; Identifying the target frequency of the target data element according to the frequency information; Determining whether the target frequency is higher than a predetermined frequency threshold; If the target frequency is not higher than the predetermined frequency threshold, determining that the type of the glass article is organic glass; If the target frequency is higher than the predetermined frequency threshold, determining that the type of the glass article is inorganic glass, and Among them, the types of glass articles also include soda-lime glass, and the step of determining the type of glass article by performing percussion sound analysis on the sound data further includes: If the target frequency is higher than a predetermined frequency threshold, determining whether the target data element corresponds to the first intensity pitch among one or more intensity pitches of the data element; If the target data element corresponds to the first intensity pitch, determining that the type of the glass article is soda-lime glass; and If the target data element does not correspond to the first intensity pitch, determining that the type of the glass article is an inorganic glass other than soda-lime glass.

11. The classification device according to claim 10, wherein the predetermined frequency threshold is 2 kHz.

12. The classification device according to claim 10, wherein the step of determining a further type of glass article by performing echo attenuation analysis on the echo data includes: Identifying the intensity information and time information of each data element of the echo data; Identifying one or more pitch data elements corresponding to one or more intensity pitches of the echo data according to the intensity information; Fitting an intensity function for each intensity of the pitch data element, thereby obtaining an attenuation coefficient of the intensity function corresponding to the echo data; Determining whether the attenuation coefficient is within a first range, a second range, or a third range; If the attenuation coefficient is within the first range, determining that the type of the glass article is soda-lime glass; If the attenuation coefficient is within the second range, determining that the type of the glass article is borosilicate glass; and If the attenuation coefficient is within the third range, determining that the type of the glass article is crystal glass.

13. The classification device according to claim 12, wherein the intensity function is presented based on the following equation: A(t) = A(0)e -B(t-C) Where A(t) represents the sound intensity as a function of time; A(0) represents the intensity of the first pitch; B represents the attenuation coefficient; and C represents the timing of the first pitch, randomly set to 0.

14. The classification device according to claim 12, wherein the requirements of the transceiver include: The diameter of the transceiver for receiving the echo is less than 3.5 mm.

15. A classification device for classifying glass articles via acoustic analysis, comprising: Sound generating means configured to generate a percussion sound by applying a percussion operation to a glass article; A transceiver electrically coupled to the sound generating means, wherein the transceiver is configured to induce an echo by applying an ultrasonic echo operation to the glass article and is further configured to receive the echo, wherein the ultrasonic echo operation is indicated by the sound generating means; A sensor configured to receive the percussion sound; and A processor electrically coupled to the sound generating means, the transceiver, and the sensor, wherein the processor is configured to execute machine instructions to implement a computer-implemented method, the method comprising: Receiving, by the processor, sound data of the percussion sound from the sensor; Through the processor, by performing percussion sound analysis on the sound data to determine the type of the glass article, where the types of the glass article include inorganic glass and organic glass; If the type of the glass article is determined to be inorganic glass, then through the processor, receive the echo data of the echo from the transceiver; and By performing sound velocity analysis on the echo data to determine a further type of the glass article, where the further types of the glass article include crystal glass, borosilicate glass, and soda-lime glass, Wherein, the step of determining the type of the glass article as organic glass, inorganic glass, or soda-lime glass by performing percussion sound analysis on the sound data includes: Identifying the intensity information and frequency information of each data element of the sound data; Identifying a target data element having the maximum intensity among all the intensities of the data elements of the sound data according to the intensity information; Identifying the target frequency of the target data element according to the frequency information; Determining whether the target frequency is higher than a predetermined frequency threshold; If the target frequency is not higher than the predetermined frequency threshold, then determine that the type of the glass article is organic glass; If the target frequency is higher than the predetermined frequency threshold, then determine that the type of the glass article is inorganic glass, and Wherein, the types of the glass article further include soda-lime glass, and the step of determining the type of the glass article by performing percussion sound analysis on the sound data further includes: If the target frequency is higher than the predetermined frequency threshold, then determine whether the target data element corresponds to the first intensity pitch among one or more intensity pitches of the data elements; If the target data element corresponds to the first intensity pitch, then determine that the type of the glass article is soda-lime glass; and If the target data element does not correspond to the first intensity pitch, then determine that the type of the glass article is an inorganic glass other than soda-lime glass.

16. The classification device according to claim 15, wherein the predetermined frequency threshold is 2 kHz.

17. The classification device according to claim 15, wherein the step of determining a further type of the glass article by performing sound velocity analysis on the echo data includes: Identifying the intensity information and time information of each data element of the echo data; Identifying two pitch data elements corresponding to two adjacent intensity pitches of the echo data according to the intensity information, and identifying the time difference between the two identified pitch data elements according to the time information; Identifying the thickness value of the wall of the glass article; Calculating the sound velocity corresponding to the echo data according to the thickness value and the identified time difference; Determining whether the calculated sound velocity is within a first range, a second range, or a third range; If the calculated sound velocity is within the first range, then determine that the type of the glass article is soda-lime glass; If the calculated sound velocity is within the second range, then determine that the type of the glass article is borosilicate glass; and If the calculated sound velocity is within the third range, then determine that the type of the glass article is crystal glass.

18. The classification device according to claim 17, wherein the speed of sound is calculated based on the following equation: where h represents the thickness value; and T d represents the time difference.

Images