Clothing material identification method and washing device
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- QINGDAO HAIER WASHING MASCH CO LTD
- Filing Date
- 2021-07-15
- Publication Date
- 2026-06-12
Smart Images
Figure CN115613265B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of smart device technology, specifically providing a method for identifying clothing material and a washing device. Background Technology
[0002] Currently, clothing is made from a variety of materials using different weaving methods and mixed with various dyes, resulting in countless combinations and significantly different washing requirements. Ignoring the washing requirements of different material combinations can easily cause severe deformation or wear and tear on clothing. Therefore, different washing processes are needed for different materials. If washing machines could automatically identify the material of the clothes and select the appropriate washing program or adjust the washing parameters accordingly, it could provide users with a more convenient and efficient laundry experience.
[0003] Currently, there are three main methods for identifying clothing materials: absorbency assessment, radio frequency identification (RFID), and image recognition. The absorbency assessment method suffers from poor detection quality and low accuracy. RFID identification requires cooperation from upstream and downstream clothing manufacturers and the supply chain, and RFID tags are expensive. Image recognition requires building a large training set of image samples and renting expensive servers for training, therefore it is usually deployed only in the cloud.
[0004] Existing technology has developed an optical method for identifying clothing materials. For example, Chinese invention patent application CN109752319A discloses an optical method and apparatus for identifying clothing materials. This method includes: irradiating the clothing to be tested with incident light; obtaining the amplitude value of the reflected light; and identifying the material of the clothing based on the amplitude value of the reflected light. When the light source irradiates the clothing, the incident light undergoes diffuse reflection on the surface of the clothing, producing reflected light. Due to the influence of the material itself, the amplitude value of the reflected light is attenuated to varying degrees compared to the amplitude value of the incident light. Therefore, the amplitude value of the reflected light carries information about the material of the clothing. By analyzing and interpreting the clothing material information carried by the amplitude value of the reflected light through a pre-set optical-clothing material standard database, the material information of the clothing is obtained, thereby identifying the material of the clothing to be tested. Since near-infrared spectra in specific wavelength bands cannot be received by a single spectral sensor receiving unit, but rather by multiple receiving units with different center wavelengths, the center wavelength of each receiving unit in the spectral sensor may differ slightly due to manufacturing processes and other factors. This directly manifests as a 10%-15% error in the intensity of the light received at the corresponding wavelength. Because the amplitude of the received spectrum varies between sensors, directly using the amplitude of reflected light to identify clothing material is not accurate enough.
[0005] Accordingly, a new technical solution is needed in this field to solve the above problems. Summary of the Invention
[0006] To address the aforementioned technical problems in the prior art, specifically the insufficient accuracy of directly using reflected light amplitude to identify clothing materials, this invention provides a method for identifying clothing materials. The method includes:
[0007] The clothing to be identified is illuminated with near-infrared light to obtain a reflectance spectrum curve with a predetermined wavelength range;
[0008] Based on the reflectance spectrum curve, a predetermined wavelength sequence and a function sequence consisting of absorbance corresponding to each wavelength are generated;
[0009] Differentiate the function sequence to determine the feature points where the derivative is zero or close to zero;
[0010] Generate a sensitive wavelength sequence consisting of wavelengths corresponding to all the aforementioned feature points;
[0011] The sensitive wavelength sequence is compared and merged with the predetermined wavelength sequence to generate the characteristic wavelength sequence of the clothing to be identified;
[0012] The characteristic wavelength sequence is compared with the standard characteristic wavelength sequence in the spectral database to determine the material of the clothing to be identified.
[0013] This clothing material identification method first determines the sensitive wavelength band of the clothing to be identified for near-infrared light, then encodes the wavelength sequence of this sensitive wavelength band to extract features, and finally compares the extracted features with the features of various materials in a database to identify the material of the clothing. Specifically, firstly, the derivative of a function sequence composed of absorbance corresponding to each wavelength is calculated to determine feature points where the derivative is zero or close to zero. Feature points with a derivative of zero or close to zero correspond to all absorption peaks and reflection peaks in the sensitive wavelength band. Absorption peaks indicate that the material has a strong absorption capacity for light of that wavelength, while reflection peaks indicate that the material has a weak absorption capacity for light of that wavelength, which are represented as troughs and peaks on the spectral curve. Therefore, the wavelengths corresponding to the troughs and peaks are the sensitive wavelengths of that material. The wavelengths corresponding to all feature points are combined to form a sensitive wavelength sequence. This sensitive wavelength sequence is then encoded to extract the material features of the clothing to be washed, i.e., it is compared and merged with a predetermined wavelength sequence to form a feature wavelength sequence of the clothing to be identified. This feature wavelength sequence is then used as an identification feature and compared with standard feature wavelength sequences corresponding to different materials in the database, thereby realizing the identification of the material of the clothing to be washed. Therefore, by adopting the above technical solution, this invention eliminates the need for any additional labels on clothing to identify clothing materials, making the identification process simpler and faster. Secondly, this invention does not directly identify materials based on the amplitude of the original reflectance spectral data, but rather on features extracted after encoding the sensitive wavelength sequence. Compared to directly identifying clothing materials based on the amplitude of reflected light, this significantly improves the accuracy or precision of the identification. Furthermore, this invention is unaffected by the output magnitude of the spectral sensor. Simultaneously, the encoding method for the characteristic wavelength sequence of the clothing to be identified is simple; it only requires focusing on the sensitive bands of the clothing material and encoding the sensitive wavelength sequence to extract the characteristic wavelength sequence, thereby completing the clothing material comparison and identification. This method involves low computational load and low energy consumption.
[0014] In the preferred embodiment of the above-mentioned method for identifying clothing materials, the differentiation of the function sequence includes: taking the first derivative of the function sequence. By taking the first derivative of the function sequence composed of absorbance corresponding to each wavelength, all absorption peaks and reflection peaks in the sensitive band can be determined, i.e., the wavelength positions where the derivative is zero.
[0015] In the preferred embodiment of the above-mentioned clothing material identification method, differentiating the function sequence includes: calculating the first and second derivatives of the function sequence. Calculating the second derivative can increase the number of effective feature points, thereby further improving the identification accuracy.
[0016] In a preferred embodiment of the above-described clothing material identification method, the method further includes determining additional feature points adjacent to each of the aforementioned feature points, and the sensitive wavelength sequence consists of wavelengths corresponding to all the aforementioned feature points and the additional feature points. Adding additional feature points near the feature points increases the tolerance of the identification; the more additional feature points there are, the greater the tolerance or error rate of the identification result, thereby improving the identification accuracy.
[0017] In a preferred embodiment of the above-described method for identifying clothing materials, comparing and merging the sensitive wavelength sequence with the predetermined wavelength sequence to generate a characteristic wavelength sequence for the clothing to be identified includes: retaining wavelengths in the predetermined wavelength sequence that are identical to each sensitive wavelength in the sensitive wavelength sequence, and setting all wavelengths that are not identical to the sensitive wavelengths to zero to obtain the characteristic wavelength sequence. The characteristic wavelength sequence generated through this zero-coding method only reflects all wavelengths of near-infrared light that the garment to be washed is sensitive to.
[0018] In a preferred embodiment of the above-described method for identifying clothing materials, comparing the characteristic wavelength sequence of the clothing to be identified with a standard characteristic wavelength sequence in a spectral database to determine the material of the clothing further includes: setting a confidence threshold; when the comparison result between the characteristic wavelength sequence of the clothing to be identified and the standard characteristic wavelength sequence in the spectral database exceeds the confidence threshold, the clothing is determined to belong to the material represented by the standard characteristic wavelength sequence. The higher the confidence threshold, the higher the identification accuracy. Therefore, the accuracy of the identification result can be controlled by the confidence threshold.
[0019] In a preferred embodiment of the above-described method for identifying clothing materials, when the comparison result between the characteristic wavelength sequence of the clothing to be identified and the standard characteristic wavelength sequence in the spectral database does not exceed the confidence threshold, the clothing to be identified is irradiated with the near-infrared spectral sensor for re-identification, or the material of the clothing to be identified is determined based on the standard characteristic wavelength sequence in the spectral database that has the highest similarity to the characteristic wavelength sequence. There are two reasons why the comparison result does not exceed the confidence threshold. The first reason is that the above identification process is incorrect; in this case, the identification can be performed again. The second reason is that the confidence threshold is set too high; in this case, the material of the clothing to be identified can be determined based on the standard characteristic wavelength sequence in the spectral database that has the highest similarity to the characteristic wavelength sequence of the clothing to be washed.
[0020] In the preferred embodiment of the above-mentioned clothing material identification method, the predetermined wavelength range is 750nm-1050nm. For a given wavelength range, different clothing materials correspond to different spectral characteristics, which can reduce interference from clothing color and improve the accuracy of clothing material identification.
[0021] In a preferred embodiment of the above-described method for identifying clothing materials, the characteristic wavelength sequence of the clothing to be identified is input into the spectral database. Inputting the encoded characteristic wavelength sequence into the spectral database as input to a machine learning classification algorithm can significantly improve recognition accuracy.
[0022] To address the aforementioned problems in the prior art, specifically the insufficient accuracy of directly using reflected light amplitude to identify clothing materials, this invention also provides a washing device. The washing device employs any of the clothing identification methods described above to identify the material of the garment to be washed. By using any of the clothing material identification methods described above, the washing device of this invention can more accurately and automatically identify the material of the garment to be washed, thereby selecting a suitable washing program or making corresponding adjustments to the washing parameters, thus providing users with a more satisfactory, convenient, and faster washing experience. Attached Figure Description
[0023] The preferred embodiments of the present invention are described below with reference to the accompanying drawings, in which:
[0024] Figure 1 This is a flowchart of the clothing material identification method of the present invention;
[0025] Figure 2 This is a flowchart of the first embodiment of the clothing material identification method of the present invention;
[0026] Figure 3 This is an example reflectance spectrum curve for a wavelength range of 750nm-1050nm;
[0027] Figure 4 These are example first and second derivative curves of the absorbance function sequence corresponding to wavelength ranges of 750nm-1050nm;
[0028] Figure 5 This is a flowchart of the second embodiment of the clothing material identification method of the present invention. Detailed Implementation
[0029] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.
[0030] To address the issue of insufficient accuracy in existing technologies that directly use reflected light amplitude to identify clothing materials, this invention provides a method for identifying clothing materials. This method includes:
[0031] The clothing to be identified is illuminated with near-infrared light to obtain a reflectance spectrum curve with a predetermined wavelength range (step S1);
[0032] Based on the reflectance spectrum curve, a predetermined wavelength sequence and a function sequence consisting of absorbance corresponding to each wavelength are generated (step S2);
[0033] Differentiate the function sequence to determine the characteristic points where the derivative is zero or close to zero (step S3);
[0034] Generate a sensitive wavelength sequence consisting of wavelengths corresponding to all feature points (step S4);
[0035] The sensitive wavelength sequence is compared and merged with the predetermined wavelength sequence to generate the characteristic wavelength sequence of the clothing to be identified (step S5);
[0036] The characteristic wavelength sequence is compared with the standard characteristic wavelength sequence in the spectral database to determine the material of the clothing to be identified (step S6).
[0037] Figure 1 This is a flowchart of the clothing material identification method of the present invention. For example... Figure 1 As shown, the clothing material identification method, after starting, first performs step S1, which involves illuminating the clothing to be identified with near-infrared light to obtain a reflectance spectrum curve with a predetermined wavelength range. The near-infrared light can be provided by a near-infrared spectral sensor. Near-infrared spectral sensors can directly detect specific components of a target object using the near-infrared spectrum. These sensors operate on the principle that these components selectively absorb specific wavelengths of near-infrared energy. In one or more embodiments, the predetermined wavelength range is 750nm-1050nm. The light source on the sensor chip emits a spectrum covering wavelengths of 750nm-1050nm with consistent intensity, illuminating the clothing to be identified, and the receiving unit receives the reflected spectral intensity. Alternatively, the predetermined wavelength range may be a portion of 750nm-1050nm. Next, the clothing material identification method performs step S2, which generates a predetermined wavelength sequence and a function sequence consisting of absorbance corresponding to each wavelength based on the reflectance spectrum curve. For example, a predetermined wavelength sequence X is denoted as: X = [750, 751, ..., 1049, 1050]. In this embodiment, the wavelengths are 300 discrete wavelength values that differ from each other by 1 nm. Therefore, the absorbance function sequence Y is denoted as: Y = [y0, y1, ..., y 300The absorbance function sequence Y consists of 300 absorbance values corresponding to 300 wavelengths. After obtaining the absorbance function sequence Y, the clothing material identification method performs step S3, which involves differentiating the absorbance function sequence Y and marking feature points where the derivative is zero or close to zero. In one or more embodiments, a predetermined number of additional feature points adjacent to each feature point are also marked to increase the recognition tolerance. In one or more embodiments, the first derivative of the absorbance function sequence Y is calculated. Alternatively, the first and second derivatives of the absorbance function sequence Y are calculated. Alternatively, more derivatives of the absorbance function sequence Y are calculated as needed. In one or more embodiments, along the horizontal axis that increases sequentially with wavelength, the additional feature points consist of two adjacent points to the left of each feature point (corresponding to two adjacent wavelengths) and two adjacent points to the right of each feature point (corresponding to two adjacent wavelengths, i.e., two wavelengths 1 nm apart), wherein the feature point is one wavelength away from its nearest neighbor, i.e., 1 nm. Therefore, each feature point has four additional feature points. Alternatively, along the horizontal axis that increases sequentially with wavelength, the additional feature points consist of one adjacent point to the left of each feature point (corresponding to two adjacent wavelengths) and one adjacent point to the right of each feature point. Therefore, each feature point has two additional feature points.
[0038] like Figure 1 As shown, after determining all feature points, the clothing material identification method executes step S4 to generate a sensitive wavelength sequence X1 composed of wavelengths corresponding to all feature points. When additional feature points are marked, the sensitive wavelength sequence X1 also includes the wavelengths corresponding to those additional feature points. For example, the sensitive wavelength sequence X1 is denoted as: X1 = [766, 767, 768, 769, 770, ...]. Next, the clothing material identification method executes step S5 to compare and merge the sensitive wavelength sequence X1 with a predetermined wavelength sequence X to generate a feature wavelength sequence of the clothing to be identified. In one or more embodiments, wavelengths in the predetermined wavelength sequence X that are identical to all sensitive wavelengths in the sensitive wavelength sequence X1 are retained, and wavelengths that are not identical to the sensitive wavelengths are all set to zero to obtain the feature wavelength sequence X2. For example, the feature wavelength sequence X2 is denoted as: X2 = [0, 0, ..., 0, 766, 767, 768, 769, 770, ...]. After obtaining the characteristic wavelength sequence X2, the clothing material identification method executes step 6, comparing the characteristic wavelength sequence X2 with a standard characteristic wavelength sequence in the spectral database to determine the material of the clothing to be identified. In one or more embodiments, a confidence threshold is first set. When the comparison result between the characteristic wavelength sequence X2 of the clothing to be identified and the standard characteristic wavelength sequence in the spectral database exceeds the confidence threshold, it is determined that the clothing to be identified belongs to the material represented by the standard characteristic wavelength sequence. After executing step 6, the clothing material identification method ends. The spectral database can be a cloud database or an offline database.
[0039] Figure 2 This is a flowchart of the first embodiment of the clothing material identification method of the present invention. Figure 2 As shown, after the clothing material identification method starts, it first executes step S1, which involves illuminating the clothing to be identified using a near-infrared spectral sensor to obtain a reflectance spectral curve with a predetermined wavelength range. Specifically, the light source on the sensor chip emits a spectrum covering wavelengths of 750nm-1050nm with consistent intensity to illuminate the clothing to be identified. The receiving unit receives the reflected spectral intensity, and then the spectral intensity data is processed by the processor to form a reflectance spectral curve. Figure 3 This is an example reflectance spectrum curve for a wavelength range of 750nm-1050nm. Figure 3 As shown in the reflectance spectrum curve, the horizontal axis represents discrete wavelengths from 750nm to 1050nm, with adjacent wavelengths differing by 1nm; the vertical axis represents the absorbance corresponding to each wavelength, with the maximum absorbance less than 0.08. Total reflectance is 1, and total absorption is 0. After obtaining the reflectance spectrum curve, the clothing material identification method then executes step S2, which generates a predetermined wavelength sequence X = [750, 751, ..., 1049, 1050] and a function sequence Y = [y0, y1, ..., y2] composed of the absorbance corresponding to each wavelength, based on the reflectance spectrum curve. 300 Then, the clothing material identification method proceeds to step S3, where the first and second derivatives of the absorbance function sequence Y are calculated, and all feature points with derivatives of zero or close to zero and four additional feature points adjacent to each feature point are marked. Figure 4 These are example first and second derivative curves of the absorbance function sequence corresponding to wavelengths ranging from 750 nm to 1050 nm. For example... Figure 4 As shown, curve 1 is the first derivative curve, and curve 2 is the second derivative curve. All characteristic points where the derivative is zero or close to zero are marked with circles on curves 1 and 2, and the wavelength corresponding to each characteristic point is plotted on the horizontal axis. Since the wavelength values are discrete values differing by 1 nm, some points where the derivative is zero do not correspond to integer wavelengths. In such cases, the integer wavelength closest to the characteristic point where the derivative is zero is selected; the point corresponding to this integer wavelength is called the near-zero characteristic point.
[0040] like Figure 2 As shown, after determining the wavelengths corresponding to all feature points and additional feature points, the clothing material recognition method proceeds to step S4, generating a sensitive wavelength sequence composed of the wavelengths corresponding to all feature points and additional feature points. For example, as... Figure 4As shown, the first derivative is close to 0 at the feature point corresponding to a wavelength of 768 nm. Therefore, the wavelengths corresponding to the four additional specific points are 766 nm, 767 nm, 769 nm, and 770 nm, respectively, thus generating a sensitive wavelength sequence X1 = [766, 767, 768, 769, 770, ...]. The second derivative is processed in the same way. The wavelengths of all specific points and additional feature points are combined to form the final sensitive wavelength sequence. After obtaining the sensitive wavelength sequence, the clothing material recognition method executes step S5, retaining wavelengths in the predetermined wavelength sequence X that are the same as each sensitive wavelength in the sensitive wavelength sequence, and setting all wavelengths that are different from the sensitive wavelengths to zero to obtain the feature wavelength sequence. For example, following the encoding method described above, the sensitive wavelength sequence X1 = [766,767,768,769,770,…] is compared and merged with the predetermined wavelength sequence X = [750,751,…,1049,1050] to obtain the feature wavelength sequence X2 = [0,0,…,0,766,767,768,769,770,…]. After obtaining the feature wavelength sequence X2 = [0,0,…,0,766,767,768,769,770,…], the clothing material identification method executes step S61, comparing the feature wavelength sequence with the standard feature wavelength sequences corresponding to different clothing materials in the database, and then determining whether the comparison result of the feature wavelength sequence with any standard feature wavelength sequence in the spectral database exceeds the confidence threshold. The confidence threshold is, for example, 90%, 95%, or 98%, or other suitable values. The confidence threshold is related to the number of recognition types and the recognition accuracy; the more recognition types and the higher the recognition accuracy, the higher the confidence threshold should be. If the judgment result is yes, it means that the comparison result of the feature wavelength sequence with a certain standard feature wavelength sequence in the spectral database exceeds the confidence threshold. Then, the clothing material recognition method executes step S62 to determine that the clothing to be identified belongs to the material represented by the corresponding standard feature wavelength sequence. After step S62 is completed, the recognition method ends. If the judgment result is no, it means that the comparison result of the feature wavelength sequence with any standard feature wavelength sequence in the spectral database does not exceed the confidence threshold. Then, the clothing material recognition method executes step S7 to determine whether step S1 has been executed twice. Step S1 involves using a near-infrared spectral sensor to irradiate the clothing to be identified to obtain a reflectance spectrum curve with a predetermined wavelength range. If the judgment result is yes, it means that the clothing to be washed has undergone the same material recognition process twice, which eliminates the possibility of an error in the recognition process. Therefore, in this case, the clothing material recognition method executes step S63 to determine the material of the clothing to be identified based on the standard feature sequence in the spectral database that has the highest similarity to the feature wavelength sequence. After step S63 is completed, the identification method ends.If the judgment result is negative, it means that the clothes to be washed have only undergone one material identification process, and the possibility of identification error cannot be ruled out. Then, step S1 is executed again, that is, the clothes to be identified are irradiated by a near-infrared spectral sensor to obtain a reflectance spectrum curve with a predetermined wavelength range, and the above identification process is repeated.
[0041] Figure 5 This is a flowchart of the second embodiment of the clothing material identification method of the present invention. Figure 5 As shown, after the clothing material identification method starts, it first executes steps S1-S5. Steps S1, S2, S3, S4, and S5 are the same as steps S1-S5 in the above embodiment, and will not be repeated here. Figure 5 As shown, after obtaining the characteristic wavelength sequence of the garment to be washed in step S5, the garment material identification method executes step S61, comparing the characteristic wavelength sequence with the standard characteristic wavelength sequences corresponding to different garment materials in the database, and then determining whether the comparison result of the characteristic wavelength sequence with any standard characteristic wavelength sequence in the spectral database exceeds a confidence threshold. For example, the confidence threshold is 95% or 98%. If the determination result is yes, it means that the comparison result of the characteristic wavelength sequence with a certain standard characteristic wavelength sequence in the spectral database exceeds the confidence threshold, and the garment material identification method executes step S62 to determine that the garment to be identified belongs to the material represented by the corresponding standard characteristic wavelength sequence. After step S62 is completed, the identification method ends. If the determination result is no, it means that the comparison result of the characteristic wavelength sequence with any standard characteristic wavelength sequence in the spectral database does not exceed the confidence threshold, and the garment material identification method executes step S7, determining whether step S1 has been executed twice. Step S1 involves using a near-infrared spectral sensor to irradiate the garment to be identified to obtain a reflectance spectrum curve with a predetermined wavelength range. If the judgment result is yes, it means that the garment to be washed has already undergone the same material identification process twice, ruling out the possibility of an error in the identification process. Therefore, in this case, the garment material identification method executes step S63, determining the material of the garment to be identified based on the standard feature sequence with the highest similarity to the feature wavelength sequence in the spectral database. After step S63 is completed, the garment material identification method proceeds to step S8, inputting the feature wavelength sequence of the garment to be identified into the spectral database as input for the machine learning classification algorithm, thereby significantly improving the identification accuracy. After step S8 is completed, the garment material identification method ends. If the judgment result is no, it means that the garment to be washed has only undergone one material identification process, and the possibility of an error in identification cannot be ruled out. In this case, step S1 is executed again, that is, the garment to be identified is irradiated with a near-infrared spectral sensor to obtain a reflectance spectrum curve with a predetermined wavelength range, and the identification process is repeated.
[0042] This invention also relates to a washing device. This washing device (not shown in the figures) includes, but is not limited to, a drum washing machine or a top-loading washing machine. This washing device uses the clothing material identification method of this invention to automatically identify the material of the clothes to be washed, and then selects a reasonable washing program or makes corresponding adjustments to the washing parameters based on the material of the clothes, which can bring users a more convenient and faster washing experience.
[0043] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after such changes or substitutions will all fall within the scope of protection of the present invention.
Claims
1. A method for identifying clothing material, characterized in that, The method for identifying clothing material includes: Near-infrared light is used to illuminate the clothing to be identified to obtain a reflectance spectrum curve with a predetermined wavelength range; Based on the reflectance spectrum curve, a predetermined wavelength sequence and a function sequence consisting of absorbance corresponding to each wavelength are generated; Differentiate the function sequence to determine the feature points where the derivative is zero or close to zero; Generate a sensitive wavelength sequence consisting of wavelengths corresponding to all the aforementioned feature points; The sensitive wavelength sequence is compared and merged with the predetermined wavelength sequence to generate the characteristic wavelength sequence of the clothing to be identified; The characteristic wavelength sequence is compared with the standard characteristic wavelength sequence in the spectral database to determine the material of the clothing to be identified.
2. The method for identifying clothing material according to claim 1, characterized in that, Differentiating the sequence of functions includes: Find the first derivative of the given function sequence.
3. The method for identifying clothing material according to claim 1, characterized in that, Differentiating the sequence of functions includes: Find the first and second derivatives of the given function sequence.
4. The method for identifying clothing material according to any one of claims 1-3, characterized in that, The clothing material identification method further includes determining additional feature points adjacent to each of the feature points, and the sensitive wavelength sequence consists of wavelengths corresponding to all the feature points and additional feature points.
5. The method for identifying clothing material according to any one of claims 1-3, characterized in that, The process of comparing and merging the sensitive wavelength sequence with the predetermined wavelength sequence to generate the characteristic wavelength sequence of the clothing to be identified includes: The wavelengths in the predetermined wavelength sequence that are the same as each sensitive wavelength in the sensitive wavelength sequence are retained, and all wavelengths that are not the same as the sensitive wavelengths are set to zero to obtain the characteristic wavelength sequence.
6. The method for identifying clothing material according to any one of claims 1-3, characterized in that, The process of comparing the characteristic wavelength sequence of the garment to be identified with a standard characteristic wavelength sequence in a spectral database to determine the material of the garment also includes: Set a confidence threshold; When the comparison result between the characteristic wavelength sequence of the garment to be identified and the standard characteristic wavelength sequence in the spectral database exceeds the confidence threshold, the garment is determined to belong to the material represented by the standard characteristic wavelength sequence.
7. The method for identifying clothing material according to claim 6, characterized in that, When the comparison result between the characteristic wavelength sequence of the garment to be identified and the standard characteristic wavelength sequence in the spectral database does not exceed the confidence threshold, the garment to be identified is irradiated with the near-infrared light for re-identification, or the material of the garment to be identified is determined based on the standard characteristic wavelength sequence in the spectral database that has the highest similarity to the characteristic wavelength sequence.
8. The method for identifying clothing material according to any one of claims 1-3, characterized in that, The predetermined wavelength range is 750nm-1050nm.
9. The method for identifying clothing material according to any one of claims 1-3, characterized in that, The characteristic wavelength sequence of the garment to be identified is input into the spectral database.
10. A washing device, characterized in that, The washing equipment uses the clothing material identification method according to any one of claims 1-9 to identify the material of the clothes to be washed.