A method and device for identifying clothing status, electronic device and storage medium

Abstract

The present application relates to a method and device for identifying clothing status, an electronic device and a storage medium; wherein the method comprises: obtaining image information of target clothing in a drum; pre-identifying the image information to obtain a pre-identification result of the target clothing; determining a target rotation speed of the drum, wherein the target rotation speed is the rotation speed of the drum when the image information is obtained; and correcting the pre-identification result according to the target rotation speed to obtain a final identification result. The method provided in the embodiment of the present application can correct the pre-identification result according to the target rotation speed of the drum, so as to accurately distinguish the clothing status that cannot be accurately distinguished in the pre-identification result, thereby achieving the purpose of improving the accuracy of clothing status identification of clothing inside the drum.

Description

Technical Field

[0001] The present application relates to the field of image recognition technology, and in particular to a clothing status recognition method and device, an electronic device, and a storage medium. Background Art

[0002] During the development of washing machine products, the washing machine needs to dry the clothes after washing. During the drying process, the washing machine drum rotates, and then hot air is introduced into the drum to dry the clothes in the drum. If the clothes are unfolded in the drum during the drying process, the clothes will not be wrinkled after drying. Therefore, it is necessary to control the speed of the drum and cooperate with the gravity of the clothes to unfold the clothes in the drum. However, the unfolding of clothes requires reasonable control of the drum speed, and controlling the speed of the drum requires knowing the state of the clothes in the drum. There are four states of clothes in the drum: rolling, falling, unfolding and sticking to the wall. Therefore, it is necessary to use a camera to capture the scene inside the drum and then identify the state of the clothes.

[0003] In the related art, since the two states of rolling and falling are very similar and difficult to distinguish during the actual algorithm implementation process, there is a low accuracy rate in identifying the state of the clothes in the drum.

[0004] With respect to the technical problem of low accuracy in identifying the state of clothes in a drum existing in the related art, no effective solution has been provided so far. Summary of the invention

[0005] In order to solve the technical problem of low accuracy in identifying the status of clothes in the drum, the present application provides a method and device for identifying the status of clothes, an electronic device and a storage medium.

[0006] In a first aspect, an embodiment of the present application provides a method for identifying a clothing state, comprising:

[0007] Acquire image information of target clothes in the drum;

[0008] Pre-recognizing the image information to obtain a pre-recognition result of the target clothing;

[0009] Determining a target rotation speed of the drum, wherein the target rotation speed is the rotation speed of the drum when acquiring the image information;

[0010] The pre-recognition result is corrected according to the target speed to obtain a final recognition result.

[0011] Optionally, as in the aforementioned method, the pre-recognition of the image information to obtain the pre-recognition result of the target clothing includes:

[0012] The image information is input into a clothing state recognition model to obtain the pre-recognition result of the target clothing, wherein the pre-recognition result includes: a one-to-one correspondence between clothing states and pre-recognition weights, and the pre-recognition weights are used to indicate the probability that the state of the target clothing is a corresponding predicted state.

[0013] Optionally, as in the aforementioned method, the step of correcting the pre-recognition result according to the target speed to obtain a final recognition result includes:

[0014] Generate a corresponding target correction function according to the target speed;

[0015] The pre-recognition result is corrected by the target correction function to obtain a final recognition result.

[0016] Optionally, as in the aforementioned method, the pre-recognition result includes: a one-to-one correspondence between a clothing state and a pre-recognition weight, the pre-recognition weight is used to indicate the probability that the state of the target clothing is a corresponding predicted state, and generating a corresponding target correction function according to the target speed includes:

[0017] Determine at least two to-be-corrected clothing states that require parameter correction among all clothing states;

[0018] Determine a weighted value uniquely corresponding to each of the clothing states to be corrected according to the target rotation speed;

[0019] The target correction function is obtained according to each of the weighted values.

[0020] Optionally, as in the aforementioned method, the clothing state to be corrected includes: a first clothing state to be corrected and a second clothing state to be corrected, and the determining, according to the target rotation speed, a weighted value uniquely corresponding to each clothing state to be corrected includes:

[0021] The first weighted value g corresponding to the first clothing state to be corrected and the second weighted value s corresponding to the second clothing state to be corrected are calculated by the following formula:

[0022]

[0023] Wherein, e is the base of the natural logarithm, r is the target rotation speed, C is a constant term, the first clothing state to be corrected is a rolling state, and the second clothing state to be corrected is a falling state.

[0024] Optionally, as in the aforementioned method, the step of correcting the pre-recognition result by using the target correction function to obtain the final recognition result includes:

[0025] For each of the clothing states to be corrected, a weight to be corrected corresponding to the clothing state to be corrected is determined in the pre-recognition result, and the weight to be corrected is weighted by the weight value corresponding to the clothing state to be corrected in the target correction function to obtain a corrected output value corresponding to the clothing state to be corrected, wherein the weight to be corrected is the pre-recognition weight corresponding to the clothing state to be corrected;

[0026] The clothing state to be corrected corresponding to the largest target corrected output value among all the corrected output values ​​is taken as the final recognition result.

[0027] Optionally, as in the aforementioned method, the pre-recognition result includes: a one-to-one correspondence between a clothing state and a pre-recognition weight, the pre-recognition weight being used to indicate a probability that the state of the target clothing is a corresponding predicted state, and after pre-recognizing the image information to obtain the pre-recognition result of the target clothing, the method further includes:

[0028] Determine a target clothing state among all clothing states that does not require parameter correction;

[0029] When the largest target pre-recognition weight among all the pre-recognition weights is the pre-recognition weight of the target clothing state, the target clothing state is taken as the final recognition result.

[0030] In a second aspect, an embodiment of the present application provides a clothing status recognition device, comprising:

[0031] An acquisition module, used to acquire image information of target clothes in the drum;

[0032] A pre-recognition module, used to pre-recognize the image information to obtain a pre-recognition result of the target clothing;

[0033] a determination module, configured to determine a target rotation speed of the drum, wherein the target rotation speed is the rotation speed of the drum when acquiring the image information;

[0034] The result module is used to correct the pre-recognition result according to the target speed to obtain a final recognition result.

[0035] In a third aspect, an embodiment of the present application provides an electronic device, comprising: a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other via the communication bus;

[0036] The memory is used to store computer programs;

[0037] The processor is used to implement any of the methods described above when executing the computer program.

[0038] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, wherein the storage medium includes a stored program, wherein the program executes a method as described in any of the preceding items when executed.

[0039] The above technical solution provided by the embodiment of the present application has the following advantages compared with the prior art:

[0040] The method provided in the embodiment of the present application can correct the pre-recognition result according to the target rotation speed of the drum, so as to accurately distinguish the clothing status that cannot be accurately distinguished in the pre-recognition result, thereby achieving the purpose of improving the accuracy of clothing status recognition of clothes under the drum. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0042] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0043] Figure 1 A schematic diagram of a method for identifying clothing status provided in an embodiment of the present application;

[0044] Figure 2 A schematic flow chart of a method for identifying clothing status provided in another embodiment of the present application;

[0045] Figure 3 A schematic flow chart of a method for identifying clothing status provided in another embodiment of the present application;

[0046] Figure 4 A block diagram of a clothing status recognition device provided in an embodiment of the present application;

[0047] Figure 5 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0048] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0049] According to one aspect of an embodiment of the present application, a method for identifying a clothing state is provided. Optionally, in this embodiment, the above-mentioned clothing state identification method can be applied to a hardware environment consisting of a terminal and a server. The server is connected to the terminal via a network, and can be used to provide clothing state identification services for the terminal or a client installed on the terminal. A database can be set on the server or independently of the server to provide data storage services for the server.

[0050] The above network may include but is not limited to at least one of the following: wired network, wireless network. The above wired network may include but is not limited to at least one of the following: wide area network, metropolitan area network, local area network, and the above wireless network may include but is not limited to at least one of the following: WIFI (Wireless Fidelity), Bluetooth. The terminal may not be limited to a PC, a mobile phone, a tablet computer, etc.

[0051] The clothing state recognition method of the embodiment of the present application can be executed by a server, a terminal, or a server and a terminal together. The terminal can also execute the clothing state recognition method of the embodiment of the present application by a client installed thereon.

[0052] Taking the method for identifying the clothing status in this embodiment executed by a terminal as an example, Figure 1 A method for identifying clothing status provided in an embodiment of the present application includes the following steps:

[0053] Step S101, acquiring image information of target clothing in the drum.

[0054] The clothing state recognition method in this embodiment can be applied to scenarios where the clothing state needs to be recognized in a device for drying clothing, such as: a scenario where clothing is spun dry in a washing machine drum, a scenario where clothing is dried in a washing machine drum, a scenario where clothing is dried in a dryer, etc., or other scenarios for drying clothing. In the embodiment of this application, the above-mentioned clothing state recognition method is described by taking the drying of clothing in a washing machine as an example. For other types of scenarios where the clothing state needs to be obtained, the above-mentioned clothing state recognition method is also applicable if there is no contradiction.

[0055] Taking the scenario of drying clothes in the drum of a washing machine as an example, the pre-recognition results of the clothes are corrected in combination with the rotation speed of the drum to improve the recognition accuracy of the clothing status.

[0056] The image inside the drum can be captured by an image capture device (for example, a camera) disposed inside the drum, thereby achieving the purpose of obtaining image information of the target clothing in the drum.

[0057] Step S102, pre-recognize the image information to obtain a pre-recognition result of the target clothing.

[0058] After the image information is acquired, the image information may be pre-identified to obtain a pre-identification result of the target clothing.

[0059] The pre-recognition may be to input the image information into a pre-trained clothing state recognition model, and then obtain a pre-recognition result for indicating the clothing state of the target clothing.

[0060] Furthermore, the clothing state recognition model may be unable to accurately distinguish certain similar clothing states. Therefore, when the clothing state is one of the similar clothing states, the pre-recognition result may have a large error.

[0061] Step S103, determining a target rotation speed of the drum, wherein the target rotation speed is the rotation speed of the drum when acquiring the image information.

[0062] While acquiring the image information, the target rotation speed of the drum can be acquired by setting a drum rotation speed acquisition device on the drum; the target rotation speed of the drum can also be acquired by acquiring the operating status of a driving unit that controls the rotation of the drum.

[0063] Step S104, correcting the pre-recognition result according to the target speed to obtain a final recognition result.

[0064] At different rotation speeds of the drum, the clothing states of the clothes in the drum are also different. Therefore, after obtaining the target rotation speed of the drum, the pre-recognition result can be corrected based on the target rotation speed to obtain the final recognition result.

[0065] Optionally, the pre-recognition result may include a pre-recognition weight corresponding to each recognition result, and the correction of the pre-recognition result according to the target speed may be to obtain a correction value for correction according to the target speed:

[0066] The correction value can be corrected by adding or subtracting the pre-recognition weight in the pre-recognition result. For example, when the correction value corresponding to the target speed is a0, the clothing state recognition model cannot accurately distinguish between clothing state A1 (pre-recognition weight is a1) and clothing state A2 (pre-recognition weight is a2), and the pre-recognition weight a1 and the pre-recognition weight a2 are greater than the pre-recognition weights of other clothing states, and the higher the correction value corresponding to the target speed, the higher the probability that the clothing state of the target clothing is A1, then the pre-recognition result is corrected by the target speed, and the weight of clothing state A1 after correction is a1+a0, and the weight of clothing state A2 after correction is a2+1-a0. Furthermore, the clothing state corresponding to the larger value of a1+a0 and a2+1-a0 can be selected as the final recognition result.

[0067] In addition, the correction value may be corrected by weighting the pre-recognition weight in the pre-recognition result. For example, when the correction value corresponding to the target speed is a0, the clothing state recognition model cannot accurately distinguish between clothing state A1 (pre-recognition weight is a1) and clothing state A2 (pre-recognition weight is a2), and the pre-recognition weight a1 and the pre-recognition weight a2 are greater than the pre-recognition weights of other clothing states, and the higher the correction value corresponding to the target speed, the higher the probability that the clothing state of the target clothing is A1, then the pre-recognition result is corrected by the target speed, and the weight of clothing state A1 after correction is a1×a0, and the weight of clothing state A2 after correction is a2×(1-a0). Furthermore, the clothing state corresponding to the larger value of a1×a0 and a2×(1-a0) can be selected as the final recognition result.

[0068] Through the method in this embodiment, the pre-recognition result can be corrected according to the target rotation speed of the drum to accurately distinguish the clothing status that cannot be accurately distinguished in the pre-recognition result, thereby achieving the purpose of improving the accuracy of clothing status recognition of clothes under the drum.

[0069] As an optional implementation, as in the above method, the step S102 performs pre-recognition on the image information to obtain the pre-recognition result of the target clothing, including the following steps:

[0070] Step S201, input the image information into the clothing state recognition model to obtain the pre-recognition result of the target clothing, wherein the pre-recognition result includes: a one-to-one correspondence between the clothing state and the pre-recognition weight, and the pre-recognition weight is used to indicate the probability that the state of the target clothing is the corresponding predicted state.

[0071] After obtaining the image information, the image information can be input into a clothing state recognition model to obtain a pre-recognition result of the target clothing, wherein the clothing state recognition model can be obtained by training a training model with training image information and after the recognition accuracy meets the preset requirements.

[0072] After the image information is input into the clothing state recognition model, the obtained pre-recognition result includes a one-to-one correspondence between clothing states and pre-recognition weights. That is, when there are clothing states I, II, III, and IIII, there are pre-recognition weights i corresponding to clothing state I, ii corresponding to clothing state II, iii corresponding to clothing state III, and iiii corresponding to clothing state IIII; further, pre-recognition weight i+pre-recognition weight ii+pre-recognition weight iii+pre-recognition weight iiii=1.

[0073] Through the method in this embodiment, the pre-recognition result of the target clothing can be first predicted by the clothing state recognition model, thereby improving the efficiency of clothing state recognition of the target clothing.

[0074] like Figure 2 As shown, as an optional implementation, as in the above method, the pre-recognition result is corrected according to the target speed to obtain the final recognition result, which includes the following steps:

[0075] Step S301, generating a corresponding target correction function according to the target speed;

[0076] Step S302, correcting the pre-recognition result by using the target correction function to obtain the final recognition result.

[0077] After the target rotation speed is obtained, a target correction function for the user to correct the pre-recognition result may be generated based on the target rotation speed.

[0078] The target correction function may be a function for processing the pre-recognition weights of the clothing states that cannot be accurately distinguished in the pre-recognition results, so that the pre-recognition weights of the clothing states that cannot be accurately distinguished can improve the accuracy of the final recognition result after being corrected by the target correction function.

[0079] like Figure 3 As shown, as an optional implementation, as in the aforementioned method, the pre-recognition result includes: a one-to-one correspondence between the clothing state and the pre-recognition weight, the pre-recognition weight is used to indicate the probability that the state of the target clothing is the corresponding predicted state, and the step S301 generates the corresponding target correction function according to the target speed, including the following steps:

[0080] Step S401, determining at least two clothing states to be corrected that require parameter correction among all clothing states.

[0081] At least two clothing states to be corrected that cannot be accurately distinguished by the clothing state recognition model may be determined in advance from among all clothing states.

[0082] The clothing state to be corrected can be obtained by identifying different image information through the clothing state recognition model. When the probability or number of errors in identifying one clothing state P as another clothing state Q by the clothing state recognition model is greater than a preset value, the clothing state P and the clothing state Q are determined as the clothing state to be corrected. For example, when the clothing states include rolling, falling, unfolding and sticking to the wall, when 1000 preset image information are recognized by the clothing state recognition model, if there are 100 times when rolling is recognized as falling, 100 times when falling is recognized as rolling, 1 time when unfolding is recognized as other clothing states, and 1 time when sticking to the wall is recognized as other clothing states, then the number and probability of errors in identifying the rolling and falling states are much greater than the unfolding and sticking to the wall. Therefore, the two clothing states to be corrected are determined to be rolling and falling.

[0083] Step S402, determining a unique weighted value corresponding to each clothing state to be corrected according to the target rotation speed.

[0084] After the target speed is determined, a weighted value uniquely corresponding to each clothing state to be corrected can be calculated by a preset calculation scheme. That is, after the target speed is determined, a unique corresponding weighted value can be determined for each clothing state to be corrected, and for other clothing states, there is no need to calculate the corresponding weighted value.

[0085] Step S403: Obtain a target correction function according to each weighted value.

[0086] After calculating the weighted value corresponding to each clothing state to be corrected, the target correction function for weighting the weight to be corrected corresponding to each clothing state to be corrected can be constructed through all weighted values. For example, when there are two clothing states to be corrected (p, q), the weighted value corresponding to the clothing state to be corrected p is g, and the weighted value corresponding to the clothing state to be corrected q is s, then the target correction function is:

[0087]

[0088] Through the method in this embodiment, it is possible to obtain a target correction function for correcting the weight to be corrected of the clothing state to be corrected, so as to obtain a final recognition result according to the target correction function later.

[0089] As an optional implementation, as in the aforementioned method, the clothing state to be corrected includes: a first clothing state to be corrected and a second clothing state to be corrected, and the step S402 of determining the weighted value uniquely corresponding to each clothing state to be corrected according to the target speed includes the following steps:

[0090] The first weighted value g corresponding to the first clothing state to be corrected and the second weighted value s corresponding to the second clothing state to be corrected are calculated by the following formula:

[0091]

[0092] Wherein, e is the base of the natural logarithm, r is the target rotation speed, C is a constant term, the first clothing state to be corrected is the rolling state, and the second clothing state to be corrected is the falling state.

[0093] When the clothing state to be corrected only includes the first clothing state to be corrected and the second clothing state to be corrected, only the two clothing states to be corrected need to be corrected. Among them, C can be a common critical speed for controlling the drum, such as 40r / min, that is, when the speed of the drum is C, the probability that the clothing state of the clothes in the drum is a rolling state is the same as the probability that it is a falling state. Therefore, it can be obtained that when the target speed r is larger, the first weighted value g corresponding to the rolling state is larger, and conversely, when the target speed r is smaller, the second weighted value s corresponding to the falling state is larger.

[0094] Through the method in this embodiment, the first weighted value g corresponding to the rolling state and the second weighted value s corresponding to the falling state can be quickly calculated, thereby facilitating the subsequent improvement of the accuracy of the recognition of the rolling state and the falling state.

[0095] As an optional implementation, as in the above method, the step S301 corrects the pre-recognition result by using the target correction function to obtain the final recognition result, including the following steps:

[0096] Step S601, for each clothing state to be corrected, determine the weight to be corrected corresponding to the clothing state to be corrected in the pre-identification result, weight the weight to be corrected by the weight value corresponding to the clothing state to be corrected in the target correction function, and obtain the corrected output value corresponding to the clothing state to be corrected, wherein the weight to be corrected is the pre-identification weight corresponding to the clothing state to be corrected.

[0097] For each clothing state to be corrected, the weight to be corrected corresponding to the clothing state to be corrected can be determined from all the pre-identification weights in the pre-identification results, and then the weight to be corrected corresponding to the clothing state to be corrected can be obtained; then, the weight to be corrected is weighted by the weighted value corresponding to the clothing state to be corrected in the objective correction function, and the corrected output value corresponding to the clothing state to be corrected can be obtained.

[0098] For example, when the pre-recognition weights corresponding to the rolling state, the falling state, the unfolding state and the wall-attached state in the pre-recognition result are 0.35, 0.45, 0.1, and 0.1 respectively, the pre-recognition result indicates that the pre-recognition result is the falling state, so it is necessary to obtain the product of the pre-recognition result and the correction function, and then make a judgment based on the corrected value. At this time, the washing machine controller data feedback can be used to obtain the washing machine drum speed of 39r / min, and C is taken as 40r / min. The correction terms g=0.7, s=0.3 can be obtained. Substituting g=0.7, s=0.3, p=0.35, q=0.45 into the target correction function, the corrected output values ​​corresponding to the rolling state and the falling state are 0.25 and 0.14 respectively.

[0099] Step S602: The clothing state to be corrected corresponding to the largest target corrected output value among all corrected output values ​​is taken as the final recognition result.

[0100] After all the corrected output values ​​are obtained, the clothing state to be corrected as the final recognition result can be selected from all clothing states to be corrected; that is, the clothing state to be corrected corresponding to the maximum corrected output value can be used as the final recognition result. For example, based on the above embodiment, the corrected output values ​​of the rolling state and the falling state are 0.25 and 0.14, and the maximum corrected output value is 0.25, so the rolling state can be used as the final recognition result.

[0101] Through the method in this embodiment, the maximum target corrected output value can be calculated based on the target correction function, and the final recognition result can be determined based on the target corrected output value, thereby achieving the probability of improving the accuracy of the final recognition result.

[0102] As an optional implementation, as in the above method, the pre-recognition result includes: a one-to-one correspondence between the clothing state and the pre-recognition weight, and the pre-recognition weight is used to indicate the probability that the state of the target clothing is the corresponding predicted state. After the image information is pre-recognized in step S102 to obtain the pre-recognition result of the target clothing, the method further includes the following steps:

[0103] Step S701, determining a target clothing state that does not require parameter correction among all clothing states.

[0104] Step S702: When the largest target pre-recognition weight among all pre-recognition weights is the pre-recognition weight of the target clothing state, the target clothing state is taken as the final recognition result.

[0105] After all clothing states are determined, a target clothing state that does not require parameter correction among all clothing states can be determined.

[0106] The target clothing state may be a clothing state that can be accurately distinguished by the clothing state recognition model among all clothing states.

[0107] After determining the pre-recognition weight corresponding to each clothing state, the pre-recognition weight corresponding to the target clothing state can be determined, and when the pre-recognition weight of the target clothing state is the largest target pre-recognition weight among all pre-recognition weights, the target clothing state can be used as the final recognition result. For example, when the pre-recognition weights of the rolling state, the falling state, the unfolding state, and the wall-attached state are 0.12, 0.13, 0.45, and 0.3 respectively, and the unfolding state and the wall-attached state are both target clothing states, and the pre-recognition weight of the unfolding state is the largest, therefore, the unfolding state is used as the final recognition result.

[0108] Through the method in this embodiment, when the maximum pre-recognition weight is the target clothing state, the final recognition result can be obtained directly based on the pre-recognition result, without executing the relevant steps in steps S103-S104, which can effectively improve the recognition efficiency.

[0109] like Figure 4 As shown, according to an embodiment of another aspect of the present application, a clothing state recognition device is also provided, comprising:

[0110] An acquisition module 1 is used to acquire image information of target clothing in the drum;

[0111] Pre-recognition module 2, used to pre-recognize the image information and obtain the pre-recognition result of the target clothing;

[0112] A determination module 3 is used to determine a target rotation speed of the drum, wherein the target rotation speed is the rotation speed of the drum when acquiring the image information;

[0113] The result module 4 is used to correct the pre-recognition result according to the target speed to obtain the final recognition result.

[0114] Specifically, the specific process of each module in the device of the embodiment of the present invention realizing its function can be referred to the relevant description in the method embodiment, which will not be repeated here.

[0115] According to another embodiment of the present application, there is also provided an electronic device, including: Figure 5 As shown, the electronic device may include: a processor 1501 , a communication interface 1502 , a memory 1503 and a communication bus 1504 , wherein the processor 1501 , the communication interface 1502 , and the memory 1503 communicate with each other via the communication bus 1504 .

[0116] Memory 1503, used for storing computer programs;

[0117] The processor 1501 is used to implement the steps of the above method embodiment when executing the program stored in the memory 1503.

[0118] The bus mentioned in the above electronic device can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus.

[0119] The communication interface is used for communication between the above electronic device and other devices.

[0120] The memory may include a random access memory (RAM) or a non-volatile memory (NVM), such as at least one disk memory. Optionally, the memory may also be at least one storage device located away from the aforementioned processor.

[0121] The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0122] An embodiment of the present application further provides a computer-readable storage medium, the storage medium including a stored program, wherein the method steps of the above method embodiment are executed when the program is run.

[0123] It should be noted that, in this article, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0124] The foregoing is merely a specific embodiment of the present invention, which enables those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features claimed herein.

Claims

1. A method for identifying clothing status, characterized in that: include: Acquire image information of target clothes in the drum; Pre-recognizing the image information to obtain a pre-recognition result of the target clothing; Determining a target rotation speed of the drum, wherein the target rotation speed is the rotation speed of the drum when acquiring the image information; Correcting the pre-recognition result according to the target speed to obtain a final recognition result; The step of correcting the pre-recognition result according to the target speed to obtain a final recognition result includes: generating a corresponding target correction function according to the target speed; correcting the pre-recognition result by using the target correction function to obtain a final recognition result; The pre-identification result includes: a one-to-one correspondence between clothing states and pre-identification weights, the pre-identification weights are used to indicate the probability that the state of the target clothing is the corresponding predicted state, and the generating of the corresponding target correction function according to the target speed includes: determining at least two clothing states to be corrected that require parameter correction among all clothing states; determining a uniquely corresponding weighted value for each of the clothing states to be corrected according to the target speed; and obtaining the target correction function according to each of the weighted values.

2. The method according to claim 1, characterized in that The pre-recognition of the image information to obtain the pre-recognition result of the target clothing includes: The image information is input into a clothing state recognition model to obtain the pre-recognition result of the target clothing, wherein the pre-recognition result includes: a one-to-one correspondence between clothing states and pre-recognition weights, and the pre-recognition weights are used to indicate the probability that the state of the target clothing is a predicted state.

3. The method according to claim 1, characterized in that The clothing state to be corrected includes: a first clothing state to be corrected and a second clothing state to be corrected, and the determining of a unique weighted value corresponding to each clothing state to be corrected according to the target rotation speed includes: The first weighted value g corresponding to the first clothing state to be corrected and the second weighted value s corresponding to the second clothing state to be corrected are calculated by the following formula: Wherein, e is the base of the natural logarithm, r is the target rotation speed, C is a constant term, the first clothing state to be corrected is a rolling state, and the second clothing state to be corrected is a falling state.

4. The method according to claim 1, characterized in that: The correcting the pre-recognition result by the target correction function to obtain the final recognition result comprises: For each of the clothing states to be corrected, a weight to be corrected corresponding to the clothing state to be corrected is determined in the pre-recognition result, and the weight to be corrected is weighted by the weight value corresponding to the clothing state to be corrected in the target correction function to obtain a corrected output value corresponding to the clothing state to be corrected, wherein the weight to be corrected is the pre-recognition weight corresponding to the clothing state to be corrected; The clothing state to be corrected corresponding to the largest target corrected output value among all the corrected output values ​​is taken as the final recognition result.

5. The method according to claim 1, characterized in that The pre-recognition result includes: a one-to-one correspondence between a clothing state and a pre-recognition weight, wherein the pre-recognition weight is used to indicate a probability that the state of the target clothing is a corresponding predicted state. After pre-recognizing the image information to obtain the pre-recognition result of the target clothing, the method further includes: Determine a target clothing state among all clothing states that does not require parameter correction; When the largest target pre-recognition weight among all the pre-recognition weights is the pre-recognition weight of the target clothing state, the target clothing state is taken as the final recognition result.

6. A clothing status recognition device, characterized in that: include: An acquisition module, used to acquire image information of target clothes in the drum; A pre-recognition module, used to pre-recognize the image information to obtain a pre-recognition result of the target clothing; a determination module, configured to determine a target rotation speed of the drum, wherein the target rotation speed is the rotation speed of the drum when acquiring the image information; A result module, used to correct the pre-recognition result according to the target speed to obtain a final recognition result; The result module is specifically used to: generate a corresponding target correction function according to the target speed; and correct the pre-recognition result by using the target correction function to obtain a final recognition result; The pre-identification result includes: a one-to-one correspondence between clothing states and pre-identification weights, wherein the pre-identification weights are used to indicate the probability that the state of the target clothing is the corresponding predicted state, and the result module is further used to: determine at least two clothing states to be corrected that require parameter correction among all clothing states; determine a uniquely corresponding weighted value for each of the clothing states to be corrected according to the target rotation speed; and obtain the target correction function according to each of the weighted values.

7. An electronic device, characterized in that: include: A processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus; The memory is used to store computer programs; The processor is used to implement the method according to any one of claims 1 to 5 when executing the computer program.

8. A computer-readable storage medium, characterized in that: The storage medium includes a stored program, wherein the program executes the method according to any one of claims 1 to 5 when executed.

Images