Gas flow information prediction system of gas meter

Abstract

The invention relates to a gas flow information prediction system for a gas meter, and the system comprises an information extraction part which is used for employing a convolutional neural network model to intelligently analyze the predicted value of the gas consumption corresponding to the fixed time segment of the current day of a single residence; and the gas quantity judgment component is used for sending a recharging reminding signal when the gas quantity value stored by the gas card associated with the current single residence is smaller than the intelligently analyzed predicted value of the gas consumption corresponding to the current single residence in the fixed time segment of the day. According to the invention, the artificial intelligence model can be adopted to intelligently analyze the predicted value of the gas consumption corresponding to the fixed time segment of the current day of the single residence; and when the value of the gas volume stored by the gas card associated with the current single residence is smaller than the predicted value of the gas usage amount corresponding to the current single residence in the fixed time segment of the day through intelligent analysis, a recharging reminding signal is sent out, so that frequent recharging is avoided, and sufficient gas usage of the single residence is ensured.

Description

Technical Field

[0001] The present invention relates to the field of gas meters, and more particularly to a gas volume information prediction system for a gas meter. Background Art

[0002] A gas meter is a device used to measure and record the usage of gas (such as natural gas, liquefied petroleum gas, etc.). Only a small glass window on the outside of the gas meter can see a roller with digits. There are also seven digits on the roller, with the first four digits in black before the decimal point and the last three digits in red. The fuels used for daily cooking by people have changed from conventional energy sources such as firewood and coal, which are seriously wasteful and polluting, to clean energy sources such as natural gas and coal gas, and even electricity. Here, the gas meter has to play its role. Its automatic cumulative function enables those who use natural gas or piped gas to easily know how much gas has been used, so as to pay the fee according to the number of cubic meters of gas consumed per month.

[0003] However, the gas volumes used by different residences within the same time segment are different. If the same margin is provided for the same time segment of different residences, it is obvious that there will be insufficient gas volume in some residences and excessive gas volume in some residences. Therefore, how to effectively predict the gas consumption of each residence in the future time segment, so as to ensure sufficient gas supply for a single residence while avoiding frequent recharging, is one of the technical problems that need to be solved in the field of gas meter bodies. Summary of the Invention

[0004] In order to solve the technical problems in the related fields, the present invention provides a gas volume information prediction system for a gas meter. By using an artificial intelligence model, based on the duration of a fixed time segment, various meter body information of the target intelligent gas meter bound to the current single residence, and the gas consumption volumes corresponding to each fixed time segment of the current single residence in the past days, the predicted value of the gas consumption volume corresponding to the fixed time segment of the current single residence on the current day is intelligently analyzed. When the gas volume value stored in the gas card associated with the current single residence is less than the predicted value of the gas consumption volume corresponding to the fixed time segment of the current single residence on the current day, a recharge reminder signal is sent, thus avoiding frequent recharging while ensuring sufficient gas supply for a single residence.

[0005] According to the present invention, there is provided a gas volume information prediction system for a gas meter, the system comprising: A target analysis mechanism for performing multiple learning on a convolutional neural network to obtain the convolutional neural network after multiple learning and output it as a convolutional neural network model, the number of times the convolutional neural network performs learning is monotonically and positively correlated with the number of permanent residents in the current single residence; The meter body acquisition mechanism is used to obtain various meter body information of the target intelligent gas meter bound to the current single-family residence. The various meter body information of the target intelligent gas meter bound to the current single-family residence are the ignition temperature, combustion speed, upper limit threshold of ignition concentration, and lower limit threshold of ignition concentration of the target intelligent gas meter bound to the current single-family residence; The simultaneous capture mechanism is used to obtain each portion of gas consumption corresponding to different fixed time segments of each past day of the current single-family residence. The number of past days is directly proportional to the usage days of the target intelligent gas meter; The information extraction component is respectively connected to the target parsing mechanism, the meter body acquisition mechanism, and the simultaneous capture mechanism, and is used to use the convolutional neural network model to intelligently analyze the predicted value of the gas consumption corresponding to the fixed time segment of the current single-family residence on the current day based on the duration of the fixed time segment, various meter body information of the target intelligent gas meter bound to the current single-family residence, and each portion of gas consumption corresponding to different fixed time segments of each past day of the current single-family residence; The gas volume judgment component is connected to the information extraction component, and is used to send a recharge reminder signal when the gas volume value stored in the gas card associated with the current single-family residence is less than the predicted value of the gas consumption corresponding to the fixed time segment of the current single-family residence on the current day analyzed by intelligence; Wherein, the gas volume judgment component is further used to send a sufficient gas volume signal when the gas volume value stored in the gas card associated with the current single-family residence is greater than or equal to the predicted value of the gas consumption corresponding to the fixed time segment of the current single-family residence on the current day analyzed by intelligence; Wherein, the simultaneous capture mechanism is used to obtain each portion of gas consumption corresponding to different fixed time segments of each past day of the current single-family residence. The number of past days is directly proportional to the usage days of the target intelligent gas meter, including: the past days and the current day form a complete time interval on the time axis; Wherein, the target parsing mechanism is used to perform multiple learning on the convolutional neural network to obtain the convolutional neural network after completing multiple learning and output it as the convolutional neural network model. The number of times the convolutional neural network performs learning is monotonically and positively correlated with the number of permanent residents in the current single-family residence, including: using a content mapping formula to represent the content mapping relationship of the monotonic positive correlation between the number of times the convolutional neural network performs learning and the number of permanent residents in the current single-family residence.

[0006] Thus, the present invention at least has the following main inventive concepts: First: Perform multiple learning on the convolutional neural network to obtain the convolutional neural network after completing multiple learning and output it as the convolutional neural network model. The number of times the convolutional neural network performs learning is monotonically and positively correlated with the number of permanent residents in the current single-family residence, so as to customize different convolutional neural network models for different single-family residences to intelligently predict the gas consumption value in the subsequent time interval; Second: Obtain various meter body information of the target intelligent gas meter bound to the current single-family residence. The various meter body information of the target intelligent gas meter bound to the current single-family residence is the ignition temperature, combustion speed, upper threshold value of ignition concentration, and lower threshold value of ignition concentration of the target intelligent gas meter bound to the current single-family residence. Also, obtain the gas consumption for each fixed time segment corresponding to each day in the past of the current single-family residence. The number of days in the past is directly proportional to the usage days of the target intelligent gas meter, thereby providing multiple basic information for the intelligent prediction of gas consumption values in subsequent time intervals; Third: Based on the duration of the fixed time segment, various meter body information of the target intelligent gas meter bound to the current single-family residence, and the gas consumption for each fixed time segment corresponding to each day in the past of the current single-family residence, intelligently analyze the predicted value of the gas consumption corresponding to the fixed time segment of the current single-family residence on the current day. When the gas volume value stored in the gas card associated with the current single-family residence is less than the predicted value of the gas consumption corresponding to the fixed time segment of the current single-family residence analyzed by intelligence, send a recharge reminder signal, thereby avoiding frequent recharges while ensuring sufficient gas supply for the single-family residence. Description of the Drawings

[0007] The embodiments of the present invention will be described below in conjunction with the drawings, where:

[0008] Figure 1 It is a structural block diagram of a gas volume information prediction system of a gas meter shown according to the primary embodiment of the present invention.

[0009] Figure 2 It is a structural block diagram of a gas volume information prediction system of a gas meter shown according to the secondary embodiment of the present invention.

[0010] Figure 3 It is a structural block diagram of a gas volume information prediction system of a gas meter shown according to the tertiary embodiment of the present invention. Detailed Embodiment

[0011] The embodiments of the gas volume information prediction system of the gas meter of the present invention will be described in detail below with reference to the drawings.

[0012] Figure 1 It is a structural block diagram of a gas volume information prediction system of a gas meter shown according to the primary embodiment of the present invention. The system includes: A target parsing mechanism for performing multiple learning on a convolutional neural network to obtain the convolutional neural network after multiple learning and output it as a convolutional neural network model. The number of times the convolutional neural network performs learning is monotonically and positively correlated with the number of permanent residents in the current single-family residence; The meter body acquisition mechanism is used to obtain various meter body information of the target intelligent gas meter bound to the current single-family residence. The various meter body information of the target intelligent gas meter bound to the current single-family residence are the ignition temperature, combustion speed, upper limit threshold of ignition concentration, and lower limit threshold of ignition concentration of the target intelligent gas meter bound to the current single-family residence; Exemplarily, the meter body acquisition mechanism is used to obtain various meter body information of the target intelligent gas meter bound to the current single-family residence. The various meter body information of the target intelligent gas meter bound to the current single-family residence are the ignition temperature, combustion speed, upper limit threshold of ignition concentration, and lower limit threshold of ignition concentration of the target intelligent gas meter bound to the current single-family residence, including: The meter body acquisition mechanism includes multiple information acquisition components for respectively acquiring the ignition temperature, combustion speed, upper limit threshold of ignition concentration, and lower limit threshold of ignition concentration of the target intelligent gas meter; The simultaneous capture mechanism is used to obtain each gas consumption corresponding to different fixed time segments of each past day of the current single-family residence. The number of days of each past day is directly proportional to the usage days of the target intelligent gas meter; The information extraction component is respectively connected to the target analysis mechanism, the meter body acquisition mechanism, and the simultaneous capture mechanism, and is used to use the convolutional neural network model to intelligently analyze the predicted value of the gas consumption corresponding to the fixed time segment of the current single-family residence on the current day based on the duration of the fixed time segment, various meter body information of the target intelligent gas meter bound to the current single-family residence, and each gas consumption corresponding to different fixed time segments of each past day of the current single-family residence; The gas volume judgment component is connected to the information extraction component, and is used to send a recharge reminder signal when the gas volume value stored in the gas card associated with the current single-family residence is less than the predicted value of the gas consumption corresponding to the fixed time segment of the current single-family residence on the current day analyzed by intelligence; Wherein, the gas volume judgment component is further used to send a sufficient gas volume signal when the gas volume value stored in the gas card associated with the current single-family residence is greater than or equal to the predicted value of the gas consumption corresponding to the fixed time segment of the current single-family residence on the current day analyzed by intelligence; Wherein, the simultaneous capture mechanism is used to obtain each gas consumption corresponding to different fixed time segments of each past day of the current single-family residence. The number of days of each past day is directly proportional to the usage days of the target intelligent gas meter, including: Each past day and the current day form a complete time interval on the time axis; Wherein, the target analysis mechanism is used to perform multiple learning on the convolutional neural network to obtain the convolutional neural network after completing multiple learning and output it as the convolutional neural network model. The number of times the convolutional neural network performs learning is monotonically and positively correlated with the number of permanent residents in the current single-family residence, including: Using a content mapping formula to represent the content mapping relationship of the monotonic positive correlation between the number of times the convolutional neural network performs learning and the number of permanent residents in the current single-family residence; Among them, obtain various body information of the target intelligent gas meter bound to the current single-family residence. The various body information of the target intelligent gas meter bound to the current single-family residence is the ignition temperature, combustion speed, upper threshold value of ignition concentration, and lower threshold value of ignition concentration of the target intelligent gas meter bound to the current single-family residence, including: the target intelligent gas meter serves the current single-family residence; And among them, using the convolutional neural network model, based on the duration of the fixed time segment, various body information of the target intelligent gas meter bound to the current single-family residence, and each gas consumption corresponding to the fixed time segment of each past day of the current single-family residence, intelligently analyze the predicted value of the gas consumption corresponding to the fixed time segment of the current single-family residence on the current day, including: parallelly inputting the duration of the fixed time segment, various body information of the target intelligent gas meter bound to the current single-family residence, and each gas consumption corresponding to the fixed time segment of each past day of the current single-family residence into the convolutional neural network model.

[0013] Figure 2 It is a structural block diagram of a gas volume information prediction system of a gas meter shown according to a secondary embodiment of the present invention.

[0014] Compared with Figure 1 , Figure 2 The gas volume information prediction system in The on-site control interface is arranged near the information extraction component, the target parsing mechanism, the body collection mechanism, and the simultaneous capture mechanism and is respectively connected to the information extraction component, the target parsing mechanism, the body collection mechanism, and the simultaneous capture mechanism; Among them, the on-site control interface is arranged near the information extraction component, the target parsing mechanism, the body collection mechanism, and the simultaneous capture mechanism and is respectively connected to the information extraction component, the target parsing mechanism, the body collection mechanism, and the simultaneous capture mechanism, including: the on-site control interface is used to respectively realize the synchronous drive control of two-by-two devices of the information extraction component, the target parsing mechanism, the body collection mechanism, and the simultaneous capture mechanism.

[0015] Figure 3 It is a structural block diagram of a gas volume information prediction system of a gas meter shown according to a further secondary embodiment of the present invention.

[0016] Compared with Figure 1 , Figure 3 The gas volume information prediction system in A parameter service mechanism, which is arranged near the information extraction component, the target parsing mechanism, the table body acquisition mechanism, and the simultaneous capture mechanism and is respectively connected to the information extraction component, the target parsing mechanism, the table body acquisition mechanism, and the simultaneous capture mechanism; Among them, the parameter service mechanism, which is arranged near the information extraction component, the target parsing mechanism, the table body acquisition mechanism, and the simultaneous capture mechanism and is respectively connected to the information extraction component, the target parsing mechanism, the table body acquisition mechanism, and the simultaneous capture mechanism, includes: the parameter service mechanism is used to provide the respective required working current values for the information extraction component, the target parsing mechanism, the table body acquisition mechanism, and the simultaneous capture mechanism.

[0017] Next, the specific structure of the gas volume information prediction system of the gas meter according to the present invention will be further described.

[0018] In the gas volume information prediction system of the gas meter according to various embodiments of the present invention: An FPGA chip is used to perform image data processing on the output data of the information extraction component, the target parsing mechanism, the table body acquisition mechanism, and the simultaneous capture mechanism to obtain the output processing data corresponding to the information extraction component, the target parsing mechanism, the table body acquisition mechanism, and the simultaneous capture mechanism respectively; Among them, using an FPGA chip to perform image data processing on the output data of the information extraction component, the target parsing mechanism, the table body acquisition mechanism, and the simultaneous capture mechanism to obtain the output processing data corresponding to the information extraction component, the target parsing mechanism, the table body acquisition mechanism, and the simultaneous capture mechanism respectively includes: performing maximum value filtering processing on the output data of the information extraction component, the target parsing mechanism, the table body acquisition mechanism, and the simultaneous capture mechanism to obtain the output processing data corresponding to the information extraction component, the target parsing mechanism, the table body acquisition mechanism, and the simultaneous capture mechanism respectively; Among them, using an FPGA chip to perform image data processing on the output data of the information extraction component, the target parsing mechanism, the table body acquisition mechanism, and the simultaneous capture mechanism to obtain the output processing data corresponding to the information extraction component, the target parsing mechanism, the table body acquisition mechanism, and the simultaneous capture mechanism respectively includes: performing minimum value filtering processing on the output data of the information extraction component, the target parsing mechanism, the table body acquisition mechanism, and the simultaneous capture mechanism to obtain the output processing data corresponding to the information extraction component, the target parsing mechanism, the table body acquisition mechanism, and the simultaneous capture mechanism respectively; Among them, using an FPGA chip to perform image data processing on the output data of the information extraction component, the target analysis mechanism, the body acquisition mechanism and the simultaneous capture mechanism to obtain the output processing data corresponding to the information extraction component, the target analysis mechanism, the body acquisition mechanism and the simultaneous capture mechanism respectively includes: performing median filtering processing on the output data of the information extraction component, the target analysis mechanism, the body acquisition mechanism and the simultaneous capture mechanism to obtain the output processing data corresponding to the information extraction component, the target analysis mechanism, the body acquisition mechanism and the simultaneous capture mechanism respectively; And wherein, an FPGA chip is used to perform image data processing on the output data of the information extraction component, the target analysis mechanism, the body acquisition mechanism and the simultaneous capture mechanism to obtain the output processing data corresponding to the information extraction component, the target analysis mechanism, the body acquisition mechanism and the simultaneous capture mechanism respectively, including: performing edge sharpening processing on the output data of the information extraction component, the target analysis mechanism, the body acquisition mechanism and the simultaneous capture mechanism to obtain the output processing data corresponding to the information extraction component, the target analysis mechanism, the body acquisition mechanism and the simultaneous capture mechanism respectively.

[0019] In addition, in the gas volume information prediction system of the gas meter, the convolutional neural network model is used to intelligently analyze the gas usage corresponding to the fixed time segment of the current single house on the day based on the duration of the fixed time segment, the various meter body information of the target smart gas meter bound to the current single house, and the gas usage corresponding to the fixed time segments of the current single house in the past days. The predicted value of the gas usage corresponding to the fixed time segment of the current single house on the day also includes: running the convolutional neural network model to obtain the predicted value of the gas usage corresponding to the fixed time segment of the current single house on the day output by the convolutional neural network model.

[0020] The gas volume information prediction system of the gas meter of the present invention solves the technical problem that the prior art cannot effectively predict the gas volume used in future time segments of each residence, making it difficult to configure the gas balance in advance. By using an artificial intelligence model to intelligently analyze the predicted value of the gas volume used corresponding to the fixed time segments of the current single residence on the day, and when the gas volume value of the gas card associated with the current single residence is less than the predicted value of the gas volume used corresponding to the fixed time segments of the current single residence on the day determined by the intelligent analysis, a recharge reminder signal is issued, thereby avoiding frequent recharges while ensuring sufficient gas consumption in the single residence, thereby solving the above technical problems.

[0021] Although the preferred embodiments of the present invention have been described in detail above, it is understood that many changes and / or improvements to the basic inventive concept taught herein will be apparent to those skilled in the art and fall within the spirit and scope of the present invention as defined by the appended claims.

Claims

1. A gas meter gas volume information prediction system, characterized in that: The system comprises: The target parsing mechanism is used to perform multiple learning on the convolutional neural network to obtain the convolutional neural network after completing the multiple learning and output it as a convolutional neural network model, wherein the number of times the convolutional neural network performs learning is monotonically positively correlated with the number of permanent residents of the current single-building residence, including: using a content mapping formula to represent the content mapping relationship of the monotonically positive correlation between the number of times the convolutional neural network performs learning and the number of permanent residents of the current single-building residence; The meter body collection mechanism is used to obtain various meter body information of the target smart gas meter bound to the current single-building house, and the various meter body information of the target smart gas meter bound to the current single-building house is the ignition temperature, combustion speed, ignition concentration upper limit threshold and ignition concentration lower limit threshold of the target smart gas meter bound to the current single-building house; At the same time, the capturing mechanism is used to obtain the gas usage corresponding to the fixed time segments of the past days of the current single house, and the number of days of the past days is proportional to the number of days of use of the target smart gas meter, including: the past days and the current day form a complete time interval on the time axis; The information extraction component is connected to the target analysis mechanism, the meter body collection mechanism and the simultaneous capture mechanism respectively, and is used to use the convolutional neural network model to intelligently analyze the predicted value of the gas usage corresponding to the fixed time segment of the current single house on the current day based on the duration of the fixed time segment, the various meter body information of the target smart gas meter bound to the current single house, and the gas usage corresponding to the fixed time segments of the current single house in the past days; A gas volume determination component connected to the information extraction component is used to send a recharge reminder signal when the gas volume value of the gas card associated with the current single-family house is less than the predicted value of the gas volume used by the current single-family house at a fixed time segment on the day according to the intelligent analysis; The gas volume judgment component is also used to send a gas volume sufficient signal when the gas volume value of the gas card associated with the current single-family house is greater than or equal to the predicted value of the gas volume used corresponding to the fixed time segment of the current single-family house on the same day based on the intelligent analysis.

2. The gas meter gas volume information prediction system according to claim 1, characterized in that: Acquire various meter body information of the target smart gas meter bound to the current single-building house, wherein the various meter body information of the target smart gas meter bound to the current single-building house is the ignition temperature, the combustion speed, the ignition concentration upper limit threshold and the ignition concentration lower limit threshold of the target smart gas meter bound to the current single-building house, including: the target smart gas meter serves the current single-building house; Among them, the convolutional neural network model is used to intelligently analyze the predicted value of the gas usage corresponding to the fixed time segment of the current single house on the day based on the duration of the fixed time segment, the various meter body information of the target smart gas meter bound to the current single house, and the various gas usage corresponding to the fixed time segments of the current single house in the past days. The value includes: the duration of the fixed time segment, the various meter body information of the target smart gas meter bound to the current single house, and the various gas usage corresponding to the fixed time segments of the current single house in the past days are input into the convolutional neural network model in parallel.

3. The gas meter gas volume information prediction system according to claim 2, characterized in that: The system further comprises: A field control interface is arranged near the information extraction component, the target analysis mechanism, the body collection mechanism and the simultaneous capture mechanism and is connected to the information extraction component, the target analysis mechanism, the body collection mechanism and the simultaneous capture mechanism respectively; Among them, the field control interface is arranged near the information extraction component, the target analysis mechanism, the body collection mechanism and the simultaneous capture mechanism and is respectively connected to the information extraction component, the target analysis mechanism, the body collection mechanism and the simultaneous capture mechanism, including: the field control interface is used to respectively realize the synchronous drive control of each of the information extraction component, the target analysis mechanism, the body collection mechanism and the simultaneous capture mechanism.

4. The gas meter gas volume information prediction system according to claim 2, characterized in that: The system further comprises: A parameter service mechanism, which is arranged near the information extraction component, the target analysis mechanism, the body collection mechanism and the simultaneous capture mechanism and is connected to the information extraction component, the target analysis mechanism, the body collection mechanism and the simultaneous capture mechanism respectively; Among them, the parameter service mechanism is arranged near the information extraction component, the target analysis mechanism, the meter body collection mechanism and the simultaneous capture mechanism and is respectively connected to the information extraction component, the target analysis mechanism, the meter body collection mechanism and the simultaneous capture mechanism, including: the parameter service mechanism is used to provide the information extraction component, the target analysis mechanism, the meter body collection mechanism and the simultaneous capture mechanism with their respective required working current values.

5. The gas volume information prediction system of a gas meter according to any one of claims 2 to 4, characterized in that: An FPGA chip is used to perform image data processing on the output data of the information extraction component, the target analysis mechanism, the body acquisition mechanism and the simultaneous capture mechanism to obtain output processing data corresponding to the information extraction component, the target analysis mechanism, the body acquisition mechanism and the simultaneous capture mechanism respectively.

6. The gas meter gas volume information prediction system according to claim 5, characterized in that: Using an FPGA chip to perform image data processing on the output data of the information extraction component, the target analysis mechanism, the body acquisition mechanism and the simultaneous capture mechanism to obtain the output processing data corresponding to the information extraction component, the target analysis mechanism, the body acquisition mechanism and the simultaneous capture mechanism respectively includes: performing maximum value filtering processing on the output data of the information extraction component, the target analysis mechanism, the body acquisition mechanism and the simultaneous capture mechanism to obtain the output processing data corresponding to the information extraction component, the target analysis mechanism, the body acquisition mechanism and the simultaneous capture mechanism respectively.

7. The gas meter gas volume information prediction system according to claim 5, characterized in that: Using an FPGA chip to perform image data processing on the output data of the information extraction component, the target analysis mechanism, the body acquisition mechanism and the simultaneous capture mechanism to obtain the output processing data corresponding to the information extraction component, the target analysis mechanism, the body acquisition mechanism and the simultaneous capture mechanism respectively includes: performing minimum value filtering processing on the output data of the information extraction component, the target analysis mechanism, the body acquisition mechanism and the simultaneous capture mechanism to obtain the output processing data corresponding to the information extraction component, the target analysis mechanism, the body acquisition mechanism and the simultaneous capture mechanism respectively.

8. The gas meter gas volume information prediction system according to claim 5, characterized in that: Using an FPGA chip to perform image data processing on the output data of the information extraction component, the target analysis mechanism, the body acquisition mechanism and the simultaneous capture mechanism to obtain the output processing data corresponding to the information extraction component, the target analysis mechanism, the body acquisition mechanism and the simultaneous capture mechanism respectively includes: performing median filtering processing on the output data of the information extraction component, the target analysis mechanism, the body acquisition mechanism and the simultaneous capture mechanism to obtain the output processing data corresponding to the information extraction component, the target analysis mechanism, the body acquisition mechanism and the simultaneous capture mechanism respectively.

9. The gas meter gas volume information prediction system according to claim 5, characterized in that: Using an FPGA chip to perform image data processing on the output data of the information extraction component, the target analysis mechanism, the body acquisition mechanism and the simultaneous capture mechanism to obtain the output processing data corresponding to the information extraction component, the target analysis mechanism, the body acquisition mechanism and the simultaneous capture mechanism respectively includes: performing edge sharpening processing on the output data of the information extraction component, the target analysis mechanism, the body acquisition mechanism and the simultaneous capture mechanism to obtain the output processing data corresponding to the information extraction component, the target analysis mechanism, the body acquisition mechanism and the simultaneous capture mechanism respectively.

Images