Water supply network leakage identification system and method

Abstract

The invention relates to a water supply network leakage identification system and method, and the method comprises the steps: carrying out the collection of an acoustic emission signal of a detected section of a water supply network through an acoustic emission collection device, carrying out the noise reduction and filtering of the acoustic emission signal through a signal noise reduction processing device in a variational mode decomposition and correlation test mode, obtaining a filtered acoustic emission signal, and carrying out the detection of the filtered acoustic emission signal. And finally, extracting the Mel-cepstrum coefficient characteristics from the filtered acoustic emission signals through a leakage identification device, and obtaining the leakage condition of the detected section of the water supply network according to the Mel-cepstrum coefficient characteristics through a deep neural calculation model. Due to the fact that the Mel-cepstrum coefficient characteristics of the acoustic emission signals obtained after noise reduction and filtering can accurately promote recognition and prediction of the leakage situation of the water supply network, it is ensured that the leakage situation of the detected section of the water supply network can be obtained by the deep neural calculation model, the leakage recognition precision is high, and human experience recognition is not needed.

Description

Technical Field

[0001] The present invention relates to the technical field of leakage prediction of water supply pipe networks, and in particular to a system and method for identifying leakage of water supply pipe networks. Background Art

[0002] As the artery of modern cities, water supply networks are of great significance to ensuring the stability of people's lives and the development of urban construction. With the annual expansion of urban water supply network mileage and the continuous increase in pipeline service time, the problem of pipeline leakage has become increasingly serious. According to the "China Urban and Rural Construction Statistical Yearbook" of the Ministry of Housing and Urban-Rural Development, the total urban public water supply in the country in 2021 was 63.076 billion cubic meters, and the amount of water leakage was 8.044 billion cubic meters, with a comprehensive leakage rate of 12.75%. The high leakage rate of the water supply network not only causes economic losses and waste of resources, but also leads to many problems, including secondary pollution of water quality endangering drinking water safety, low local water supply pressure affecting users' normal water use, and even causing road collapse and serious secondary disasters. Therefore, leakage control of urban water supply network has always been a problem that needs to be solved urgently in the process of urban construction.

[0003] In order to reduce the harm caused by leakage, it is necessary to explore a funny method for identifying leakage in water supply pipelines to minimize the average duration of leakage and reduce the total leakage volume.

[0004] At present, the commonly used leak detection methods in the market include night flow method and leak detection method. These methods have the characteristics of low efficiency and high labor cost, while other detection methods including pipeline robots and satellite monitoring have the problems of high cost and impact on pipeline production. At present, the leak detection method based on leakage noise signal analysis has been widely used and studied because of its reliability and wide range of applications. However, the existing domestic technologies are mainly electronic leak detectors. Most of these products are inefficient, have high false alarm rates, and require a lot of manpower and material resources. Summary of the invention

[0005] The technical problem to be solved by the present invention is how to overcome the technical defects of the existing water supply network that relies on human experience to identify and has low leakage identification accuracy. In order to overcome the above defects of the prior art, the present invention provides a water supply network leakage identification system and method, including a water supply network leakage identification system and a water supply network leakage identification method.

[0006] The present invention provides a water supply network leakage identification system, comprising:

[0007] An acoustic emission collection device collects acoustic emission signals from the tested section of the water supply network;

[0008] A signal noise reduction processing device performs noise reduction filtering processing on the acoustic emission signal by means of variational mode decomposition and correlation test to obtain a filtered acoustic emission signal;

[0009] A leakage identification device, which extracts Mel-frequency cepstral coefficient features from the filtered acoustic emission signal, and obtains leakage conditions of the measured section of the water supply network according to the Mel-frequency cepstral coefficient features through a deep neural computing model;

[0010] The signal noise reduction processing device is electrically connected to the acoustic emission collection device, and the leakage identification device is electrically connected to the signal noise reduction processing device.

[0011] The water supply network leakage identification system disclosed in the present invention collects acoustic emission signals from the tested section of the water supply network by setting an acoustic emission collection device, and then the signal noise reduction processing device performs noise reduction filtering processing on the acoustic emission signal in a variational mode decomposition and correlation test manner to obtain a filtered acoustic emission signal, and finally extracts the Mel cepstral coefficient feature from the filtered acoustic emission signal by a leakage identification device, and obtains the leakage of the tested section of the water supply network according to the Mel cepstral coefficient feature through a deep neural computing model. Since the Mel cepstral coefficient feature of the acoustic emission signal after noise reduction filtering can more accurately promote the identification and prediction of the leakage of the water supply network, it is ensured that the leakage of the tested section of the water supply network can be obtained by the deep neural computing model, and the leakage identification accuracy is high, and there is no need to rely on human experience identification, thereby avoiding the problems of relying on human experience and low accuracy in the existing leak detection work. At the same time, the system can be embedded in the online service system of the Water Conservancy Bureau, which is conducive to the monitoring and operation of the water supply network by the staff, and has certain economic significance.

[0012] In a possible implementation, the acoustic emission collection device includes:

[0013] An acoustic emission sensor converts the acoustic signal of the measured section of the water supply network into electrical signal to obtain an electrical signal;

[0014] A preamplifier is used to amplify the electrical signal to obtain an amplified signal;

[0015] A collector converts the amplified signal into a digital signal to obtain the acoustic emission signal;

[0016] in,

[0017] The preamplifier is electrically connected to the acoustic emission sensor, the collector is electrically connected to the preamplifier, and the signal noise reduction processing device is electrically connected to the collector;

[0018] This solution converts the acoustic signal of the water pipe into an electrical signal through an acoustic emission sensor, then amplifies the electrical signal through a preamplifier to obtain an amplified signal, and finally converts the amplified signal into a digital signal through a collector for easy storage and noise reduction processing, thereby ensuring the collection of acoustic emission signals in the water supply network.

[0019] In a possible implementation manner, the signal noise reduction processing device includes:

[0020] A variational mode decomposition device performs variational mode decomposition on the acoustic emission signal to obtain a plurality of eigenmodes;

[0021] A correlation testing device, performing Person correlation analysis on each of the eigenmodes and the leakage signal to obtain respective first correlation coefficients, performing Person correlation analysis on each of the eigenmodes and the noise signal to obtain respective second correlation coefficients, and then retaining the eigenmodes whose first correlation coefficients are higher than the second correlation coefficients to obtain the filtered acoustic emission signal;

[0022] in,

[0023] The variational mode decomposition device is electrically connected to the acquisition instrument, and the correlation verification device is electrically connected to the variational mode decomposition device;

[0024] This scheme first obtains multiple eigenmodes through a variational mode decomposition device, and then retains the eigenmodes with a first correlation coefficient higher than a second correlation coefficient through a correlation verification device to obtain a filtered acoustic emission signal, thereby achieving reasonable and efficient acquisition of the signal after noise reduction filtering.

[0025] In a possible implementation manner, the leakage identification device includes:

[0026] A feature extraction device extracts Mel-frequency cepstral coefficient features from the filtered acoustic emission signal;

[0027] A leakage acquisition device, which calls the deep neural computing model embedded therein to acquire the leakage condition of the measured section of the water supply network according to the Mel-frequency cepstral coefficient feature;

[0028] in,

[0029] The feature extraction device is electrically connected to the correlation inspection device, and the leakage acquisition device is electrically connected to the feature extraction device;

[0030] By proposing the Mel-frequency cepstral coefficient characteristics of the acoustic emission signal after noise reduction filtering, the leakage of the water supply network can be identified and predicted more accurately, avoiding the problems of relying on human experience and low accuracy in existing leak detection work.

[0031] In a possible implementation, the deep neural computing model includes:

[0032] A long short-term memory unit is used to obtain the temporal dependency of the Mel-frequency cepstral coefficient feature in the time dimension;

[0033] A convolution-pooling unit obtains the potential feature information of the temporal dependency through sliding convolution and pooling;

[0034] A regularization unit, performing regularization processing on the potential feature information to obtain leakage conditions of the measured section of the water supply network;

[0035] Therefore, the long short-term memory unit can be used to obtain the temporal dependency of the Mel-frequency cepstral coefficient features in the time dimension; then the convolution-pooling unit is used to perform sliding convolution on the time series to obtain the potential characteristics of the time series, and finally, through the regularization processing of the regularization unit, a water supply network leakage prediction model is constructed.

[0036] In a possible implementation, the leakage identification device further includes an alarm device, which is used to send out an alarm signal when the leakage condition of the measured section of the water supply network is leakage, and the alarm device is electrically connected to the leakage acquisition device.

[0037] Another technical solution of the present invention is to provide a method for identifying leakage in a water supply network, the method comprising the following steps:

[0038] S1: The variational mode decomposition method is optimized by the sparrow search algorithm to determine the optimal number of eigenmodes and obtain a variational mode decomposition method with optimized parameters;

[0039] S2; collecting the acoustic emission signal of the measured section of the water supply network through the acoustic emission collection device;

[0040] S3: performing variational modal decomposition on the acoustic emission signal by using the variational modal decomposition method with the parameter optimization to obtain eigenmodes with an optimal number;

[0041] S4: performing Person correlation analysis on each of the eigenmodes and the leakage signal to obtain respective first correlation coefficients, performing Person correlation analysis on each of the eigenmodes and the noise signal to obtain respective second correlation coefficients, and then retaining the eigenmode whose first correlation coefficient is higher than the second correlation coefficient to obtain a filtered acoustic emission signal;

[0042] S5: extracting Mel-frequency cepstral coefficient features from the filtered acoustic emission signal, and obtaining leakage conditions of the measured section of the water supply network according to the Mel-frequency cepstral coefficient features through a deep neural computing model.

[0043] The method disclosed in the present invention adopts a deep neural computing model after obtaining the filtered acoustic emission signal, and proposes the Mel-frequency cepstral coefficient characteristics of the acoustic emission signal after noise reduction filtering, which can more accurately identify and predict the leakage of the water supply network, avoiding the problems of relying on human experience and low accuracy in the existing leak detection work. At the same time, the method can be embedded in the online service system of the Water Conservancy Bureau, which is conducive to the staff's monitoring and operation of the water supply network, and has certain economic significance. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 This is a schematic diagram of the structure of a water supply network leakage identification system disclosed in Embodiment 1 of the present invention;

[0045] Figure 2 This is a flow chart of a method for identifying leakage in a water supply network disclosed in Embodiment 2 of the present invention. DETAILED DESCRIPTION

[0046] First, those skilled in the art should understand that these implementations are only used to explain the technical principles of the embodiments of the present application, and are not intended to limit the protection scope of the embodiments of the present application. Those skilled in the art can make adjustments to them as needed to adapt to specific application scenarios.

[0047] In the description of the embodiments of the present application, it should be noted that, unless otherwise clearly specified and limited, the terms "electrical connection" and "electrical connection relationship" should be understood in a broad sense, that is, referring to a connection method with an electrical relationship, for example, it can be a circuit connection through a conductive wire, or it can be an electrical connection through a radio signal channel (channel), or a combination of the two. In addition, "electrical connection" and "electrical connection relationship" can be based on mechanical connection (such as a conductive wire set in a connecting key); it can be a direct connection or an indirect connection through an intermediate medium. For those of ordinary skill in the art, the specific meanings of the above terms in the embodiments of the present application can be understood according to specific circumstances.

[0048] In the embodiments of the present application, unless otherwise clearly specified and limited, a first feature being "above" or "below" a second feature may mean that the first and second features are in direct contact, or the first and second features are in indirect contact through an intermediate medium. Moreover, a first feature being "above", "above" or "above" a second feature may mean that the first feature is directly above or obliquely above the second feature, or simply means that the first feature is higher in level than the second feature. A first feature being "below", "below" or "below" a second feature may mean that the first feature is directly below or obliquely below the second feature, or simply means that the first feature is lower in level than the second feature.

[0049] Two embodiments will be used below, and the present application will be further described in detail in conjunction with the accompanying drawings and specific embodiments.

[0050] Embodiment 1:

[0051] See also Figure 1 As shown, the embodiment of the present application discloses a water supply network leakage identification system. Figure 1 As a structural diagram, the system includes an acoustic emission collection device, a signal noise reduction processing device and a leakage identification device, wherein the signal noise reduction processing device is electrically connected to the acoustic emission collection device, and the leakage identification device is electrically connected to the signal noise reduction processing device.

[0052] In the system, the acoustic emission acquisition device is configured to collect the acoustic emission signals of the measured section of the water supply network; the signal noise reduction processing device is configured to perform noise reduction filtering processing on the acoustic emission signals by means of variational mode decomposition and correlation test to obtain filtered acoustic emission signals; the leakage identification device is configured to extract Mel cepstral coefficient features from the filtered acoustic emission signals, and obtain the leakage situation of the measured section of the water supply network according to the Mel cepstral coefficient features through a deep neural computing model;

[0053] Please continue to see Figure 1 As shown, the acoustic emission acquisition device with a sampling frequency of 3MHz is used in this embodiment, including an acoustic emission sensor, a preamplifier and a collector, wherein the preamplifier is electrically connected to the acoustic emission sensor, the collector is electrically connected to the preamplifier, and the signal noise reduction processing device is electrically connected to the collector. The acoustic emission sensor is configured to convert the acoustic signal of the measured section of the water supply network into an acoustic-electrical conversion to obtain an electrical signal; the preamplifier is configured to amplify the electrical signal to obtain an amplified signal; the collector is configured to convert the amplified signal into a digital signal to obtain an acoustic emission signal.

[0054] Please continue to see Figure 1 As shown, in this embodiment, the signal denoising processing device includes a variational mode decomposition device and a correlation checking device, wherein the variational mode decomposition device is electrically connected to the acquisition instrument, and the correlation checking device is electrically connected to the variational mode decomposition device. The variational mode decomposition device is configured to perform variational mode decomposition on the acoustic emission signal to obtain a plurality of eigenmodes; the correlation checking device is configured to perform Person correlation analysis on each eigenmode and the leakage signal to obtain a respective first correlation coefficient, and perform Person correlation analysis on each eigenmode and the noise signal to obtain a respective second correlation coefficient, and then retain the eigenmode whose first correlation coefficient is higher than the second correlation coefficient to obtain the filtered acoustic emission signal.

[0055] Please continue to see Figure 1As shown, in this embodiment, the leakage identification device includes a feature extraction device, a leakage acquisition device and an alarm device, wherein the feature extraction device is electrically connected to the correlation inspection device, the leakage acquisition device is electrically connected to the feature extraction device, and the alarm device is electrically connected to the leakage acquisition device. The feature extraction device is configured to extract the Mel cepstral coefficient feature from the filtered acoustic emission signal; the leakage acquisition device is configured to call the deep neural computing model embedded therein to obtain the leakage condition of the measured section of the water supply network according to the Mel cepstral coefficient feature; the alarm device is configured to send an alarm signal when the leakage condition of the measured section of the water supply network is leakage, thereby reminding the management personnel to repair the relevant pipelines.

[0056] The relevant parameters of the deep neural computing model in this embodiment are shown in the following table, including long short-term memory units, convolution-pooling units and regularization units arranged in sequence along the running direction. The long short-term memory unit is composed of two long short-term memory layers, which are used to obtain the temporal dependency of the Mel-frequency cepstral coefficient features in the time dimension; the convolution-pooling unit is composed of two convolution layers and two pooling layers, and its running order is convolution layer 1→pooling layer 1→convolution layer 2→pooling layer 2, which is used to obtain the potential feature information of the temporal dependency through sliding convolution and pooling; the regularization unit is composed of two fully connected layers, which is used to regularize the potential feature information and obtain the leakage situation of the measured section of the water supply network.

[0057]

[0058] The temporal dependency of the Mel-frequency cepstral coefficient features in the time dimension is obtained through the long short-term memory unit. Then, the convolution-pooling unit is used to perform sliding convolution on the time series to obtain the potential features of the time series. Finally, the regularization unit is used for regularization processing to form a water supply network leakage prediction model.

[0059] The water supply network leakage identification system disclosed in this embodiment is to collect the acoustic emission signal of the tested section of the water supply network by setting an acoustic emission collection device, and then the signal noise reduction processing device performs noise reduction filtering processing on the acoustic emission signal by variational mode decomposition and correlation test to obtain the filtered acoustic emission signal, and finally extract the Mel cepstral coefficient feature from the filtered acoustic emission signal by the leakage identification device, and obtain the leakage of the tested section of the water supply network according to the Mel cepstral coefficient feature through the deep neural computing model. Since the Mel cepstral coefficient feature of the acoustic emission signal after noise reduction filtering can accurately promote the identification and prediction of the leakage of the water supply network, it is ensured that the leakage of the tested section of the water supply network can be obtained by the deep neural computing model, and the leakage identification accuracy is high, without relying on human experience identification, thereby avoiding the problems of relying on human experience and low accuracy in the existing leak detection work. At the same time, the system can be embedded in the online service system of the Water Conservancy Bureau, which is conducive to the monitoring and operation of the water supply network by the staff, and has certain economic significance.

[0060] Embodiment 2:

[0061] See also Figure 2 As shown, this embodiment further discloses a method for identifying leakage in a water supply network. Figure 2 is a flow chart of the method, the method comprising the following steps:

[0062] S1: The variational mode decomposition method is optimized by the sparrow search algorithm to determine the optimal number of eigenmodes and obtain a variational mode decomposition method with optimized parameters.

[0063] Specifically, the process of optimizing the variational mode decomposition method by the sparrow search algorithm in this embodiment is as follows:

[0064] (1) Initialize the algorithm characteristic parameters of the sparrow search algorithm. The algorithm characteristic parameters include the initial number of the sparrow population, the proportion of discoverers in the sparrow population, the proportion of followers in the sparrow population, the warning value, and the safety value.

[0065] (2) Initialize the position of individual sparrows in the solution space.

[0066] (3) Calculate the envelope entropy value of each sparrow individual.

[0067] (4) Update the location of the discoverer and the individual who is aware of the danger.

[0068] (5) Calculate the envelope entropy value of each sparrow, and obtain the position information of the individual sparrow with the optimal fitness value in the solution space through multiple cycles of iteration, and output the optimal parameter combination [K, α], where K is the optimal number of eigenmodes.

[0069] S2: Collect the acoustic emission signal of the measured section of the water supply network through the acoustic emission collection device.

[0070] S3: Perform variational modal decomposition on the acoustic emission signal through the variational modal decomposition method with parameter optimization to obtain K eigenmodes.

[0071] S4: Performing Person correlation analysis on each eigenmode and the leakage signal to obtain their respective first correlation coefficients, performing Person correlation analysis on each eigenmode and the noise signal to obtain their respective second correlation coefficients, and then retaining the eigenmode whose first correlation coefficient is greater than the second correlation coefficient to obtain a filtered acoustic emission signal by reconstruction.

[0072] S5: extracting Mel-frequency cepstral coefficient features from the filtered acoustic emission signal, and obtaining leakage conditions of the measured section of the water supply network according to the Mel-frequency cepstral coefficient features through a deep neural computing model.

[0073] The method disclosed in this embodiment optimizes the variational mode decomposition method by using the sparrow search algorithm to determine the optimal number of eigenmodes, thereby obtaining a variational mode decomposition method with optimized parameters. After obtaining the filtered acoustic emission signal, a deep neural computing model is used to propose the Mel-frequency cepstral coefficient feature of the acoustic emission signal after noise reduction filtering, which can more accurately identify and predict the leakage of the water supply network, avoiding the problems of relying on human experience and low precision in the existing leak detection work. At the same time, this method can be embedded in the online service system of the Water Conservancy Bureau, which is beneficial for the staff to monitor and operate the water supply network, and has certain economic significance.

[0074] In the description of the embodiments of the present application, it should be noted that in the description of the present application, terms such as "inside" and "outside" indicating directions or positional relationships are based on the directions or positional relationships shown in the drawings. This is only for the convenience of description, and does not indicate or imply that the device or component must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it cannot be understood as a limitation on the present application.

[0075] In the description of the present application, the description with reference to the terms "one embodiment", "some embodiments", "in the present embodiment", "specific example", or "some examples" etc. means that the specific features, mechanisms, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, mechanisms, materials or characteristics described may be combined in a suitable manner in any one or more embodiments or examples. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0076] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. A water supply network leakage identification system, characterized in that: include: An acoustic emission collection device collects acoustic emission signals from the tested section of the water supply network; A signal noise reduction processing device performs noise reduction filtering processing on the acoustic emission signal by means of variational mode decomposition and correlation test to obtain a filtered acoustic emission signal; A leakage identification device, which extracts Mel-frequency cepstral coefficient features from the filtered acoustic emission signal, and obtains leakage conditions of the measured section of the water supply network according to the Mel-frequency cepstral coefficient features through a deep neural computing model; The signal noise reduction processing device is electrically connected to the acoustic emission collection device, and the leakage identification device is electrically connected to the signal noise reduction processing device.

2. The water supply network leakage identification system according to claim 1, characterized in that: The acoustic emission collection device comprises: An acoustic emission sensor converts the acoustic signal of the measured section of the water supply network into electrical signal to obtain an electrical signal; A preamplifier is used to amplify the electrical signal to obtain an amplified signal; A collector converts the amplified signal into a digital signal to obtain the acoustic emission signal; in, The preamplifier is electrically connected to the acoustic emission sensor, the collector is electrically connected to the preamplifier, and the signal noise reduction processing device is electrically connected to the collector.

3. The water supply network leakage identification system according to claim 2, characterized in that: The signal noise reduction processing device comprises: A variational mode decomposition device performs variational mode decomposition on the acoustic emission signal to obtain a plurality of eigenmodes; A correlation testing device, performing Person correlation analysis on each of the eigenmodes and the leakage signal to obtain respective first correlation coefficients, performing Person correlation analysis on each of the eigenmodes and the noise signal to obtain respective second correlation coefficients, and then retaining the eigenmodes whose first correlation coefficients are higher than the second correlation coefficients to obtain the filtered acoustic emission signal; in, The variational mode decomposition device is electrically connected to the acquisition instrument, and the correlation verification device is electrically connected to the variational mode decomposition device.

4. The water supply network leakage identification system according to claim 3, characterized in that: The leakage identification device comprises: A feature extraction device extracts Mel-frequency cepstral coefficient features from the filtered acoustic emission signal; A leakage acquisition device, which calls the deep neural computing model embedded therein to acquire the leakage condition of the measured section of the water supply network according to the Mel-frequency cepstral coefficient feature; in, The feature extraction device is electrically connected to the correlation verification device, and the leakage acquisition device is electrically connected to the feature extraction device.

5. The water supply network leakage identification system according to claim 4, characterized in that: The deep neural computing model includes: A long short-term memory unit is used to obtain the temporal dependency of the Mel-frequency cepstral coefficient feature in the time dimension; A convolution-pooling unit obtains the potential feature information of the temporal dependency through sliding convolution and pooling; The regularization unit performs regularization processing on the potential feature information to obtain the leakage situation of the measured section of the water supply network.

6. The water supply network leakage identification system according to claim 4 or 5, characterized in that: The leakage identification device also includes an alarm device, which is used to send out an alarm signal when the leakage condition of the measured section of the water supply network is leakage, and the alarm device is electrically connected to the leakage acquisition device.

7. A method for identifying leakage in a water supply network, characterized in that: The water supply network leakage identification system according to any one of claims 1 to 6 comprises the following steps: S1: The variational mode decomposition method is optimized by the sparrow search algorithm to determine the optimal number of eigenmodes and obtain a variational mode decomposition method with optimized parameters; S2; collecting the acoustic emission signal of the measured section of the water supply network through the acoustic emission collection device; S3: performing variational modal decomposition on the acoustic emission signal by using the variational modal decomposition method with the parameter optimization to obtain eigenmodes with an optimal number; S4: performing Person correlation analysis on each of the eigenmodes and the leakage signal to obtain respective first correlation coefficients, performing Person correlation analysis on each of the eigenmodes and the noise signal to obtain respective second correlation coefficients, and then retaining the eigenmode whose first correlation coefficient is higher than the second correlation coefficient to obtain a filtered acoustic emission signal; S5: extracting Mel-frequency cepstral coefficient features from the filtered acoustic emission signal, and obtaining leakage conditions of the measured section of the water supply network according to the Mel-frequency cepstral coefficient features through a deep neural computing model.

Images