Agricultural multimedia data joint compression method, device, equipment and medium

Abstract

The present invention provides a method, device, equipment and medium for joint compression of agricultural multimedia data, which relates to the field of multimedia data processing, including: processing crop images to be compressed and environmental time series data respectively according to discrete wavelet transform DWT, obtaining all detail parameters of the crop images to be compressed in the wavelet domain and wavelet transform coefficients of environmental time series data; selecting target detail parameters corresponding to the parameter area to be replaced from the diagonal high-frequency detail parameters according to the length characteristics of the wavelet transform coefficients of the environmental time series data; inserting the wavelet transform coefficients of the environmental time series data into the parameter area to be replaced, determining the diagonal high-frequency detail parameters after replacement, and sending a compressed bit stream to a decision center. The present invention not only reduces transmission energy consumption through efficient compression performance, but also can align multi-source data associated in time and space, simplify the data processing process, and help improve the accuracy of agricultural situation deduction, thereby achieving accurate and intelligent agricultural decision-making.

Description

Technical Field

[0001] The present invention relates to the field of multimedia data processing, and in particular to a method, device, equipment and medium for joint compression of agricultural multimedia data. Background Art

[0002] Data fusion methods in the agricultural field mainly focus on single-source sensor perception data, and perform data fusion on environmental scalar data or crop phenotypic image data respectively. However, the decision-making center needs to combine multi-source heterogeneous data to make agricultural situation judgments. The single-source fusion method ignores the correlation characteristics between image data and environmental scalar data, and requires the decision-making center to align data in time and space, which increases the complexity of the algorithm and also causes error accumulation. At the same time, the single-source fusion method has limited compression effect on sensor data, and agricultural resource data lacks complete integration, which makes it impossible to obtain more timely and accurate decision support. Summary of the invention

[0003] The present invention provides a method, device, equipment and medium for joint compression of agricultural multimedia data, which is used to solve the technical problem that multiple sensor data cannot be combined in the prior art. A technical solution is adopted that uses a wavelet encoder to compress crop images and embed environmental time series data into the compressed crop images.

[0004] In a first aspect, the present invention provides a method for joint compression of agricultural multimedia data, comprising:

[0005] Process the crop image to be compressed and the environmental time series data respectively according to discrete wavelet transform DWT to obtain all detail parameters of the crop image to be compressed in the wavelet domain and the wavelet transform coefficients of the environmental time series data;

[0006] Selecting target detail parameters corresponding to the parameter area to be replaced from the diagonal high-frequency detail parameters according to the length characteristics of the wavelet transform coefficients of the environmental time series data;

[0007] Inserting the wavelet transform coefficients of the environmental time series data into the parameter area to be replaced, determining the high-frequency detail parameters in the diagonal direction after the replacement, and sending a compressed bit stream to the decision center;

[0008] The diagonal high-frequency detail parameter is determined from all the detail parameters;

[0009] The mixed crop data is determined according to the replaced diagonal high-frequency detail parameters and all the detail parameters;

[0010] The compressed bit stream is generated after compressing the mixed crop data;

[0011] The environmental time series data is determined based on environmental scalar data collected by the multimedia device, and the collection time of the environmental scalar data includes the time of collecting the crop image to be compressed, multiple time points before collecting the crop image to be compressed, and multiple time points after collecting the crop image to be compressed.

[0012] According to the agricultural multimedia data joint compression method provided by the present invention, the target detail parameters corresponding to the parameter area to be replaced are selected from the diagonal high-frequency detail parameters according to the length characteristics of the wavelet transform coefficients of the environmental time series data, including:

[0013] Constructing a spatial domain corresponding to the diagonal high-frequency detail parameter according to the diagonal high-frequency detail parameter;

[0014] Determine the target detail parameters from the spatial domain according to the wavelet transform coefficients, and determine the area where the target detail parameters are located as the parameter area to be replaced;

[0015] The length feature is a wavelet quantization signal of a preset length.

[0016] According to the agricultural multimedia data joint compression method provided by the present invention, the wavelet transform coefficients of the environmental time series data are inserted into the parameter area to be replaced, and the diagonal high-frequency detail parameters after replacement are determined, including:

[0017] Determine the maximum value of the wavelet transform coefficients in all environmental time series data, and determine the maximum value as the correction scale factor;

[0018] The replaced diagonal high-frequency detail parameter is determined according to the correction scale factor and the parameter area to be replaced.

[0019] According to the agricultural multimedia data joint compression method provided by the present invention, the crop image to be compressed and the environmental time series data are processed separately according to the discrete wavelet transform DWT to obtain all detail parameters of the crop image to be compressed in the wavelet domain and the wavelet transform coefficients of the environmental time series data, including:

[0020] Using two-dimensional discrete wavelet transform DWT to process the crop image to be compressed, and obtaining all detail parameters of the crop image to be compressed in the wavelet domain, wherein all detail parameters at least include the diagonal high-frequency detail parameters;

[0021] The environmental time series data is processed using one-dimensional discrete wavelet transform (DWT) to obtain the wavelet transform coefficients of the environmental time series data.

[0022] According to the agricultural multimedia data joint compression method provided by the present invention, the step of determining the high-frequency detail parameters of the diagonal direction after replacement and sending the compressed bit stream to the decision center includes:

[0023] Determine the mixed crop data according to the replaced diagonal high-frequency detail parameters and all the detail parameters;

[0024] A mixed crop image is generated according to the mixed crop data, and the compressed bit stream is generated after the mixed crop image is compressed, so as to send the compressed bit stream to the decision center.

[0025] In a second aspect, the present invention provides a method for joint compression of agricultural multimedia data, comprising:

[0026] receiving a compressed bit stream from a codec, and obtaining a replaced diagonal high-frequency detail parameter;

[0027] Extracting wavelet parameters of environmental time series data according to the replaced diagonal high-frequency detail parameters, and acquiring environmental time series data by using inverse discrete wavelet transform IDWT;

[0028] Determine the diagonal high-frequency detail parameter according to the replaced diagonal high-frequency detail parameter and the predicted value;

[0029] Processing all detail parameters according to inverse discrete wavelet transform IDWT to reconstruct the crop image to be compressed;

[0030] The predicted value is determined according to the adjacent coordinates of the coordinate value of the rectangular area function of the diagonal high-frequency detail parameter after the replacement;

[0031] The replaced diagonal high-frequency detail parameters are determined according to the mixed crop data and all the detail parameters;

[0032] The mixed crop data is generated after decompressing the compressed bit stream.

[0033] According to the agricultural multimedia data joint compression method provided by the present invention, the wavelet parameters of the environmental time series data are extracted according to the replaced diagonal high-frequency detail parameters, and the environmental time series data are obtained by using the inverse discrete wavelet transform IDWT, including:

[0034] Determine the wavelet parameters of the environmental time series data according to the quotient of the diagonal high-frequency detail parameter after the replacement and the correction scale factor;

[0035] The one-dimensional inverse discrete wavelet transform IDWT is used to process the wavelet parameters of the acquired environmental time series data to determine the environmental time series data.

[0036] According to the agricultural multimedia data joint compression method provided by the present invention, the diagonal high-frequency detail parameter is determined according to the replaced diagonal high-frequency detail parameter and the predicted value, comprising:

[0037] Determine a first coefficient value of a first adjacent coordinate point, a second coefficient value of a second adjacent coordinate point, and a third coefficient value of a third adjacent coordinate point according to the coordinate value of the rectangular area function in the space domain corresponding to the diagonal high-frequency detail parameter after the replacement;

[0038] When the maximum value of the first coefficient value or the second coefficient value is less than the third coefficient value, the predicted value is the minimum value of the first coefficient value or the second coefficient value;

[0039] In the case where the minimum value of the first coefficient value or the second coefficient value is greater than or equal to the third coefficient value, the predicted value is the maximum value of the first coefficient value or the second coefficient value;

[0040] In the case where the maximum value of the first coefficient value or the second coefficient value is greater than or equal to the third coefficient value, or in the case where the minimum value of the first coefficient value or the second coefficient value is less than the third coefficient value, the predicted value is determined based on the sum of the first coefficient value and the second coefficient value and based on the difference between the sum and the third coefficient value;

[0041] Among them, the first adjacent coordinate point is the left pixel coordinate value of the rectangular area function coordinate value, the second adjacent coordinate point is the upper pixel coordinate value of the rectangular area function coordinate value, and the third adjacent coordinate point is the upper left pixel coordinate value of the rectangular area function coordinate value.

[0042] According to the agricultural multimedia data joint compression method provided by the present invention, the receiving of the compressed bit stream from the codec and obtaining the replaced diagonal high-frequency detail parameters include:

[0043] generating a mixed crop image according to the compressed bit stream to generate mixed crop data according to the mixed crop image;

[0044] The replaced diagonal high-frequency detail parameters are determined according to the mixed crop data and all the detail parameters.

[0045] In a third aspect, a device for jointly compressing agricultural multimedia data is provided, comprising:

[0046] An acquisition unit is used to process the crop image to be compressed and the environmental time series data respectively according to discrete wavelet transform DWT, and acquire all detail parameters of the crop image to be compressed in the wavelet domain and the wavelet transform coefficients of the environmental time series data;

[0047] A processing unit: used for selecting target detail parameters corresponding to the parameter area to be replaced from the diagonal high-frequency detail parameters according to the length characteristics of the wavelet transform coefficients of the environmental time series data;

[0048] A sending unit: used for inserting the wavelet transform coefficients of the environmental time series data into the parameter area to be replaced, determining the high-frequency detail parameters in the diagonal direction after replacement, and sending a compressed bit stream to the decision center;

[0049] The diagonal high-frequency detail parameter is determined from all the detail parameters;

[0050] The mixed crop data is determined according to the replaced diagonal high-frequency detail parameters and all the detail parameters;

[0051] The compressed bit stream is generated after compressing the mixed crop data;

[0052] The environmental time series data is determined based on environmental scalar data collected by the multimedia device, and the collection time of the environmental scalar data includes the time of collecting the crop image to be compressed, multiple time points before collecting the crop image to be compressed, and multiple time points after collecting the crop image to be compressed.

[0053] In a fourth aspect, a device for joint compression of agricultural multimedia data is provided, comprising:

[0054] Receiving unit: used for receiving the compressed bit stream from the codec, and obtaining the high-frequency detail parameters of the diagonal direction after replacement;

[0055] An extraction unit: used for extracting the wavelet parameters of the environment time series data according to the replaced diagonal high-frequency detail parameters, and obtaining the environment time series data by using inverse discrete wavelet transform IDWT;

[0056] A determining unit: configured to determine the diagonal high-frequency detail parameter according to the replaced diagonal high-frequency detail parameter and the predicted value;

[0057] Reconstruction unit: used for processing all detail parameters according to inverse discrete wavelet transform IDWT to reconstruct the crop image to be compressed;

[0058] The predicted value is determined according to the adjacent coordinates of the coordinate values ​​of the rectangular area function of the diagonal high-frequency detail parameter after the replacement;

[0059] The replaced diagonal high-frequency detail parameters are determined according to the mixed crop data and all the detail parameters;

[0060] The mixed crop data is generated after decompressing the compressed bit stream.

[0061] In a fifth aspect, an electronic device is also provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the agricultural multimedia data joint compression method when executing the program.

[0062] In a sixth aspect, a non-transitory computer-readable storage medium is also provided, on which a computer program is stored, and when the computer program is executed by a processor, the agricultural multimedia data joint compression method is implemented.

[0063] The present invention provides a method, device, equipment and medium for joint compression of agricultural multimedia data. The present invention obtains all detail parameters of the crop image to be compressed in the wavelet domain and the wavelet transform coefficients of environmental time series data according to discrete wavelet transform DWT, determines diagonal high-frequency detail parameters from all detail parameters, selects target detail parameters corresponding to the parameter area to be replaced from the diagonal high-frequency detail parameters, inserts the wavelet transform coefficients of the environmental time series data into the parameter area to be replaced, determines the diagonal high-frequency detail parameters after replacement, determines mixed crop data according to the diagonal high-frequency detail parameters after replacement and all detail parameters, generates a compressed bit stream after compressing the mixed crop data, and sends the compressed bit stream to a decision center. The present invention uses the same encoder to compress and reconstruct agricultural spatiotemporal data of different standard formats, while maintaining good reconstruction quality and decision accuracy, it can minimize the overall data rate, reduce the energy consumption of agricultural hybrid wireless sensor networks, and extend the network life. The present invention not only reduces transmission energy consumption through efficient compression performance, but also can align multi-source data associated in time and space, simplify the data processing process, and help improve the accuracy of agricultural situation deduction, thereby achieving accurate and intelligent agricultural decision-making. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0065] Figure 1 This is one of the flow charts of the agricultural multimedia data joint compression method provided by the present invention;

[0066] Figure 2 It is a schematic diagram of a process for selecting target detail parameters provided by the present invention;

[0067] Figure 3 It is a schematic diagram of a process for determining the diagonal high-frequency detail parameters after replacement provided by the present invention;

[0068] Figure 4 It is a schematic diagram of a flow chart of obtaining all detail parameters and wavelet transform coefficients provided by the present invention;

[0069] Figure 5 This is the second flow chart of the agricultural multimedia data joint compression method provided by the present invention;

[0070] Figure 6 This is the third flow chart of the joint image processing method provided by the present invention;

[0071] Figure 7 This is the fourth flow chart of the joint image processing method provided by the present invention;

[0072] Figure 8 This is one of the structural schematic diagrams of the agricultural multimedia data joint compression device provided by the present invention;

[0073] Fig. 9 This is the second structural schematic diagram of the agricultural multimedia data joint compression device provided by the present invention;

[0074] Fig.10 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0075] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0076] The development of multimedia technology has promoted intelligent applications such as crop growth identification, disease detection, and automated environmental control in the field of precision agriculture, which has also led to a sharp increase in the amount of information stored and transmitted in the agricultural Internet of Things. In traditional agricultural Internet of Things, environmental parameters such as air temperature and humidity, soil temperature and humidity, and light are sensed and transmitted by scalar sensor nodes. The amount of data is limited, and fusion transmission through cluster heads can easily exhaust energy and cause node failure. However, scalar data fusion methods are not suitable for high-throughput multimedia crop phenotypic data processing and transmission. For agricultural Internet of Things where multimedia sensors and scalar sensors coexist, the existing inefficient and non-scalable data fusion algorithms cannot meet the requirements of heterogeneity and high-throughput characteristics of sensor data.

[0077] Intelligent decision-making models that rely on single environmental scalar data or single image multimedia data have limited effect on improving the accuracy of agricultural operation decisions. Exploring the data correlation between multimedia data and environmental scalar data and establishing a joint analysis model for multimodal data can achieve more reliable agricultural production process simulation. However, sensor data must be transmitted through low-bandwidth and high-latency communication channels. Multimodal agricultural perception data increases the transmission burden of the network.

[0078] Based on the above technical obstacles that cannot be overcome by the prior art, the present invention provides a joint image processing method. Figure 1 This is one of the flow charts of the agricultural multimedia data joint compression method provided by the present invention, the execution subject is a codec, or a sensor with a codec, including:

[0079] Process the crop image to be compressed and the environmental time series data respectively according to discrete wavelet transform DWT to obtain all detail parameters of the crop image to be compressed in the wavelet domain and the wavelet transform coefficients of the environmental time series data;

[0080] Selecting target detail parameters corresponding to the parameter area to be replaced from the diagonal high-frequency detail parameters according to the length characteristics of the wavelet transform coefficients of the environmental time series data;

[0081] Inserting the wavelet transform coefficients of the environmental time series data into the parameter area to be replaced, determining the high-frequency detail parameters in the diagonal direction after the replacement, and sending a compressed bit stream to the decision center;

[0082] The diagonal high-frequency detail parameter is determined from all the detail parameters;

[0083] The mixed crop data is determined according to the replaced diagonal high-frequency detail parameters and all the detail parameters;

[0084] The compressed bit stream is generated after compressing the mixed crop data;

[0085] The environmental time series data is determined based on environmental scalar data collected by the multimedia device, and the collection time of the environmental scalar data includes the time of collecting the crop image to be compressed, multiple time points before collecting the crop image to be compressed, and multiple time points after collecting the crop image to be compressed.

[0086] In step 101, the crop image to be compressed and the environmental time series data are processed respectively according to discrete wavelet transform DWT to obtain all detail parameters of the crop image to be compressed in the wavelet domain and the wavelet transform coefficients of the environmental time series data. The crop image to be compressed is processed according to two-dimensional discrete wavelet transform DWT to obtain all detail parameters of the crop image to be compressed in the wavelet domain, wherein the detail parameters at least include diagonal high-frequency detail parameters. The present invention aims to utilize the diagonal high-frequency detail parameters to achieve the determination of the spatial domain in the subsequent step 1 and the replacement of the wavelet transform coefficients of the environmental time series data. The environmental time series data is processed according to one-dimensional discrete wavelet transform DWT to obtain the wavelet transform coefficients of the environmental time series data. The length of the environmental time series data can be defined as N, which can be determined according to the setting of the monitoring period of the Internet of Things sensor. The length of the wavelet coefficient is determined according to the length of the environmental time series data.

[0087] In step 102, target detail parameters corresponding to the parameter area to be replaced are selected from the diagonal high-frequency detail parameters according to the length characteristics of the wavelet transform coefficients of the environmental time series data. The present invention obtains the spatial domain corresponding to the diagonal high-frequency detail parameters, determines the target detail parameters from the spatial domain, and determines the parameter area to be replaced according to the target detail parameters. The crop image to be compressed is based on crop phenotypic data obtained by multimedia sensors, which is used to monitor agricultural conditions in real time, so as to make decisions in time. Discrete wavelet transform DWT is a spectral analysis method that discretizes the scale and translation of the basic wavelet, which can examine the frequency domain characteristics of the local time domain process and the time domain characteristics of the local frequency domain process. For the crop image to be compressed, it can transform the image into a series of wavelet coefficients and efficiently compress and store these coefficients, and the rough edge of the wavelet eliminates the block effect commonly found in compression, so that the image can be better restored and expressed.

[0088] Those skilled in the art understand that, assuming that multiple media sensor nodes are deployed in the plot to be monitored, and the wireless multimedia sensor network WMSN is in a fully connected state, the base station is located outside the network area. All deployed multimedia sensor nodes are energy-constrained and have processing capabilities. These nodes cannot be moved after deployment, and the aggregation node is a high-end multimedia sensor node with computing and processing capabilities. The target detail parameter is determined by limiting the coordinate value of the rectangular area function in the spatial domain corresponding to the high-frequency detail parameter. After the crop image to be compressed is processed by discrete wavelet transform DWT, the crop image to be compressed is decomposed into different bands in space and time, and good positioning is allowed in the time and frequency domains. The parameter area to be replaced for inserting environmental parameters is determined in the spatial domain corresponding to the high-frequency detail coefficient. Specifically, the original image is decomposed by DWT to obtain all detail parameters, and the diagonal high-frequency detail parameters are selected for encoding, so that the image generated after the reconstruction of the original image is degraded into wavelet coefficients.

[0089] In step 103, the wavelet transform coefficients of the environmental time series data are inserted into the parameter area to be replaced, the diagonal high-frequency detail parameters after replacement are determined, and a compressed bit stream is sent to the decision center. The wavelet transform coefficients of the environmental time series data are inserted into the parameter area to be replaced, and the diagonal high-frequency detail parameters after replacement are determined. The environmental time series data is determined based on the environmental scalar data collected by the multimedia device. The collection time of the environmental scalar data includes the time of collecting the crop image to be compressed, multiple time points before collecting the crop image to be compressed, and multiple time points after collecting the crop image to be compressed. That is, the present invention first collects environmental scalar data at one time or multiple times according to the sensor, and the collection time interval can be set to a fixed interval or a random interval, and then the collected environmental scalar data is processed by discrete wavelet transform DWT to determine the environmental time series data, and the wavelet transform coefficients of the environmental time series data are replaced into the parameter area to be replaced, and the diagonal high-frequency detail parameters after replacement are determined.

[0090] Optionally, the mixed crop data is determined based on the replaced diagonal high-frequency detail parameters and all the detail parameters, and the compressed bit stream is generated after compressing the mixed crop data. Since all the detail parameters include not only the diagonal high-frequency detail parameters but also other detail parameters, the mixed crop data is determined based on the replaced diagonal high-frequency detail parameters and other detail parameters, and then the compressed bit stream is generated after compressing the mixed crop data.

[0091] Optionally, a compressed bit stream is sent to a decision center, where the mixed crop data is crop image data to be compressed embedded with environmental time series data. The mixed crop data is then introduced into an encoder, and a suitable bit rate is selected for compression. After the mixed crop image is compressed, the sensors located in each deployment area send the compressed bit stream to the decision center.

[0092] In an optional embodiment, the determining of the replaced diagonal high frequency detail parameters and sending the compressed bit stream to the decision center comprises:

[0093] Determine the mixed crop data according to the replaced diagonal high-frequency detail parameters and all the detail parameters;

[0094] A mixed crop image is generated according to the mixed crop data, and the compressed bit stream is generated after the mixed crop image is compressed, so as to send the compressed bit stream to the decision center.

[0095] In such an embodiment, the mixed crop data is not compressed, but first restored to a mixed crop image based on the mixed crop data. The restoration process is the inverse process of processing the crop image to be compressed according to the discrete wavelet transform DWT. After the mixed crop image is determined, the mixed crop image is compressed using the JPEG 2000 encoder to generate the compressed bit stream, so as to send the compressed bit stream to the decision center.

[0096] At present, wireless sensor networks are applied in the agricultural field, mainly for sensing, analyzing and storing high-bandwidth images, for capturing data such as air temperature and humidity, soil temperature and humidity, and light, and for analysis and storage. However, farms are vast, the climate is complex, the network nodes are large in scale, and the amount of sensed data is large, which will increase the processing and communication costs of wireless sensor networks. Compared with single sensor systems, hybrid wireless sensor systems have the advantages of broadening the sensing range and improving the perception of environmental conditions. However, multimedia data is much larger than the scalar in WSN, requiring higher transmission bandwidth and additional system resources to process. In general, the battery life of sensors is limited, and it is not feasible to increase the energy of exhausted sensor nodes during transmission. The data fusion technology adopted by the present invention eliminates the redundancy in data collected from different sources, thereby reducing the number of transmissions.

[0097] In the agricultural field, due to the dense crops, complex terrain, frequent climate changes and the temporal and spatial correlation with influencing factors such as crop location, the data fusion method in the agricultural field mainly targets single-source sensor perception data, and performs data fusion on environmental scalar data or crop phenotypic image data respectively, while the decision center needs to combine multi-source heterogeneous data to make agricultural judgments. The single-source data fusion method ignores the correlation characteristics of image data and environmental scalar data, and requires the decision center to perform data temporal and spatial alignment, which increases the complexity of the algorithm and also causes error accumulation. At the same time, single-source fusion has limited compression effect on sensor data, lacks complete integration of agricultural resource data, and cannot associate, relate and combine information from multiple sensors to obtain more timely and accurate decision support. Therefore, the present invention uses multi-source heterogeneous data to make agricultural judgments, embeds agricultural environmental parameters into crop phenotypic images, or inserts image data into agricultural environmental information, so as to establish a multi-modal and mutually correlated agricultural production database.

[0098] Environmental changes in the agricultural field are affected by crop growth status and different farming operations. Single-source data will lead to uncertain decisions. The present invention analyzes the characteristics of crop phenotypic image data and environmental scalar data and provides a sensor with a codec and a joint image processing method thereof that can jointly compress and reconstruct agricultural images and environmental data for transmitting agricultural production feature data over a wireless channel. According to the crop image acquisition time, representative environmental parameters are extracted to form a climate time series, and then the crop image is compressed using a wavelet encoder, and the environmental time series data is embedded in the compressed crop image.

[0099] The present invention processes the crop image to be compressed according to the 2D discrete wavelet transform DWT, processes the environmental monitoring time series data according to the 1D discrete wavelet transform, determines the parameter area to be replaced according to the image diagonal high frequency coefficient, inserts the environmental time series wavelet quantization signal in sequence to form the replaced diagonal high frequency detail parameter, and uses the JPEG-2000 encoder to compress and encode the mixed data (the crop image mixed wavelet transform coefficient inserted with the environmental time series wavelet quantization data) to obtain the compressed bit stream. The decision center receives the compressed bit stream, obtains the decoded mixed data by using the JPEG-2000 decoder, extracts the environmental time series wavelet quantization data, obtains the environmental monitoring time series data by using the 1D discrete wavelet inverse transform, reconstructs the diagonal high frequency detail parameter of the crop image according to the replaced diagonal high frequency detail parameter, and reconstructs the crop image according to the 2D discrete wavelet inverse transform.

[0100] In view of the massive multi-source heterogeneous sensor data in agriculture, the joint compression method of crop phenotype maps and environmental parameters provided by the present invention performs wavelet decomposition on crop phenotype images, inserts associated environmental parameters into the detail parameter part, and utilizes hierarchical tree set segmentation encoding / decoding to fuse multi-source heterogeneous agricultural production data. Then, based on the separation and reconstruction strategy of compressed data, the restored agricultural feature data provides a rich foundation for precise agricultural intelligent decision-making. The present invention requires only a small amount of data preprocessing. Unlike traditional compression technology that uses two codecs to perform image and scalar book compression separately, the spiral insertion function is used to perform the environmental insertion stage in the wavelet subband. It not only has efficient compression performance to reduce transmission energy consumption, but also can align multi-source data associated in time and space, simplifying the data processing process, and helping to improve the accuracy of agricultural situation deduction, thereby achieving precise and intelligent agricultural decision-making.

[0101] The present invention provides a method, device, equipment and medium for joint compression of agricultural multimedia data. The present invention obtains all detail parameters of the crop image to be compressed in the wavelet domain and the wavelet transform coefficients of environmental time series data according to discrete wavelet transform DWT, determines diagonal high-frequency detail parameters from all detail parameters, selects target detail parameters corresponding to the parameter area to be replaced from the diagonal high-frequency detail parameters, inserts the wavelet transform coefficients of the environmental time series data into the parameter area to be replaced, determines the diagonal high-frequency detail parameters after replacement, determines mixed crop data according to the diagonal high-frequency detail parameters after replacement and all detail parameters, generates a compressed bit stream after compressing the mixed crop data, and sends the compressed bit stream to a decision center. The present invention uses the same encoder to compress and reconstruct agricultural spatiotemporal data of different standard formats, while maintaining good reconstruction quality and decision accuracy, it can minimize the overall data rate, reduce the energy consumption of agricultural hybrid wireless sensor networks, and extend the network life. The present invention not only reduces transmission energy consumption through efficient compression performance, but also can align multi-source data associated in time and space, simplify the data processing process, and help improve the accuracy of agricultural situation deduction, thereby achieving accurate and intelligent agricultural decision-making.

[0102] Figure 2 The present invention provides a flow chart of selecting target detail parameters, wherein the target detail parameters corresponding to the parameter area to be replaced are selected from the diagonal high-frequency detail parameters according to the length characteristics of the wavelet transform coefficients of the environmental time series data, including:

[0103] Constructing a spatial domain corresponding to the diagonal high-frequency detail parameter according to the diagonal high-frequency detail parameter;

[0104] Determine the target detail parameters from the spatial domain according to the wavelet transform coefficients, and determine the area where the target detail parameters are located as the parameter area to be replaced;

[0105] The length feature is a wavelet quantization signal of a preset length.

[0106] In step 1021, the diagonal high-frequency detail parameters are high-frequency detail parameters in the diagonal direction, and the spatial domain corresponding to the diagonal high-frequency detail parameters can be expressed as a spatial domain rectangular area function (x0, y0, w, h), where x0 and y0 represent the upper left corner coordinates of the interval to be inserted, and w and h are the insertion width and height, respectively.

[0107] In step 1022, the target detail parameters are determined from the spatial domain based on the wavelet transform coefficients, and the area where the target detail parameters are located is determined as the parameter area to be replaced. Optionally, the target detail parameters are determined from the spatial domain based on the length characteristics of the wavelet transform coefficients, and the area where the target detail parameters are located is determined as the parameter area to be replaced.

[0108] Figure 3 The present invention provides a flow chart for determining the diagonal high-frequency detail parameters after replacement, wherein the wavelet transform coefficients of the environmental time series data are inserted into the parameter area to be replaced to determine the diagonal high-frequency detail parameters after replacement, including:

[0109] Determine the maximum value of the wavelet transform coefficients in all environmental time series data, and determine the maximum value as the correction scale factor;

[0110] The replaced diagonal high-frequency detail parameter is determined according to the correction scale factor and the parameter area to be replaced.

[0111] In step 1031, it is assumed that the environmental time series data is E={E1, E2,…, Ez}, and the wavelet coefficients of the environmental time series data after discrete wavelet transform are S={s1, s2,…, si,…, sN}, where si is the wavelet parameter of the i-th environmental time series data. Since the mean value of the high-frequency detail parameter is close to 0, in order to prevent the image from being seriously distorted after replacing the parameters, the present invention uses a correction scale factor to correct the environmental sample data within the range of 0 to 1. The present invention uses an image compression encoder to truncate the high-frequency coefficients. Before actually inserting at least one environmental time series data into the parameter area to be replaced and determining the high-frequency detail parameters after replacement, in order to prevent the environmental parameter samples from being affected by the quantization step in the encoder, the present invention uses a scale factor to correct the environmental parameters. The correction scale factor is used to characterize the replacement of the high-frequency detail parameters with continuous environmental time series data. Optionally, the value of the correction scale factor is 35.

[0112] In step 1032, the present invention uses a one-dimensional discrete wavelet transform to convert the environmental time series data into a wavelet quantization signal of length N. The insertion process of the i-th environmental time series data wavelet quantization sample can refer to the following formula:

[0113] HH(x'+2k,y'+2l)=α·s i ,i=0,…,N (2)

[0114]

[0115] k=0:w1-1,l=0:h1-1

[0116] In formula (2), HH represents the diagonal high-frequency detail parameter decomposed by DWT, Indicates rounding down, s i represents the i-th environmental time series data, a is the correction scale factor, HH(x'+2k,y'+2l) is the parameter area to be replaced, that is, the target detail parameter. In order to obtain a better compression effect, x0, y0, w, h can be rounded to an even number for calculation.

[0117] Figure 4 The present invention provides a flow chart of obtaining all detail parameters and wavelet transform coefficients, wherein the crop image to be compressed and the environmental time series data are processed respectively according to the discrete wavelet transform DWT to obtain all detail parameters of the crop image to be compressed in the wavelet domain and the wavelet transform coefficients of the environmental time series data, including:

[0118] Using two-dimensional discrete wavelet transform DWT to process the crop image to be compressed, and obtaining all detail parameters of the crop image to be compressed in the wavelet domain, wherein all detail parameters at least include the diagonal high-frequency detail parameters;

[0119] The environmental time series data is processed using one-dimensional discrete wavelet transform (DWT) to obtain the wavelet transform coefficients of the environmental time series data.

[0120] In step 1011, a two-dimensional discrete wavelet transform DWT is used to process the crop image to be compressed to obtain all detail parameters of the crop image to be compressed in the wavelet domain, wherein all detail parameters at least include the diagonal high-frequency detail parameters. In the present invention, it is intended to complete the determination of the diagonal high-frequency detail parameters after subsequent replacement based on the diagonal high-frequency detail parameters. However, when determining the mixed crop data, it is still necessary to determine the diagonal high-frequency detail parameters after replacement and the remaining detail parameters.

[0121] In step 1012, the environmental time series data is processed using a one-dimensional discrete wavelet transform DWT to obtain wavelet transform coefficients of the environmental time series data, wherein the wavelet transform coefficients of the environmental time series data are a sequence including length characteristics of the wavelet transform coefficients of the environmental time series data.

[0122] Figure 5 This is the second flow chart of the agricultural multimedia data joint compression method provided by the present invention, the execution subject is a decision center, including:

[0123] receiving a compressed bit stream from a codec, and obtaining a replaced diagonal high-frequency detail parameter;

[0124] Extracting wavelet parameters of environmental time series data according to the replaced diagonal high-frequency detail parameters, and acquiring environmental time series data by using inverse discrete wavelet transform IDWT;

[0125] Determine the diagonal high-frequency detail parameter according to the replaced diagonal high-frequency detail parameter and the predicted value;

[0126] Processing all detail parameters according to inverse discrete wavelet transform IDWT to reconstruct the crop image to be compressed;

[0127] The predicted value is determined according to the adjacent coordinates of the coordinate values ​​of the rectangular area function of the diagonal high-frequency detail parameter after the replacement;

[0128] The replaced diagonal high-frequency detail parameters are determined according to the mixed crop data and all the detail parameters;

[0129] The mixed crop data is generated after decompressing the compressed bit stream.

[0130] In step 201, a compressed bit stream is received from a codec, and the replaced diagonal high-frequency detail parameters are obtained. The mixed crop data is generated after decompressing the compressed bit stream, and the replaced diagonal high-frequency detail parameters are determined based on the mixed crop data and other detail parameters among all the detail parameters.

[0131] In step 202, the wavelet parameters of the environmental time series data are extracted according to the replaced diagonal high-frequency detail parameters, the environmental time series data are acquired by using the inverse discrete wavelet transform IDWT, the wavelet parameters of the environmental time series data are determined according to the quotient of the replaced diagonal high-frequency detail parameters and the correction scale factor, the wavelet parameters of the acquired environmental time series data are processed by using a one-dimensional inverse discrete wavelet transform IDWT to determine the environmental time series data.

[0132] The environmental time series data is determined by the following formula:

[0133]

[0134] Formula (3) is the inverse operation of encoding, is the environmental time series data, α is the correction scale factor, and HH(x'+2k,y'+2l) is the parameter area to be replaced, that is, the target detail parameter.

[0135] In step 203, the diagonal high-frequency detail parameter is determined according to the replaced diagonal high-frequency detail parameter and the predicted value, wherein the predicted value is determined according to the adjacent coordinates of the rectangular area function coordinate value of the replaced diagonal high-frequency detail parameter, including:

[0136] Determine a first coefficient value of a first adjacent coordinate point, a second coefficient value of a second adjacent coordinate point, and a third coefficient value of a third adjacent coordinate point according to the coordinate value of the rectangular area function in the space domain corresponding to the diagonal high-frequency detail parameter after the replacement;

[0137] When the maximum value of the first coefficient value or the second coefficient value is less than the third coefficient value, the predicted value is the minimum value of the first coefficient value or the second coefficient value;

[0138] In the case where the minimum value of the first coefficient value or the second coefficient value is greater than or equal to the third coefficient value, the predicted value is the maximum value of the first coefficient value or the second coefficient value;

[0139] In the case where the maximum value of the first coefficient value or the second coefficient value is greater than or equal to the third coefficient value, or in the case where the minimum value of the first coefficient value or the second coefficient value is less than the third coefficient value, the predicted value is determined based on the sum of the first coefficient value and the second coefficient value and based on the difference between the sum and the third coefficient value;

[0140] Among them, the first adjacent coordinate point is the left pixel coordinate value of the rectangular area function coordinate value, the second adjacent coordinate point is the upper pixel coordinate value of the rectangular area function coordinate value, and the third adjacent coordinate point is the upper left pixel coordinate value of the rectangular area function coordinate value.

[0141] Those skilled in the art understand that, in the process of encoding the environmental time series data, a portion of the HH coefficients in the decomposed image are replaced by the environmental time series data. However, in order to resolve the uncertainty of the replacement coefficients, the present invention uses a predictor for prediction and performs the following prediction before reconstructing the host image.

[0142] The median edge detection (MED) predictor is the core of the compression standard and is used for lossless compression and near-lossless compression algorithms of continuous tone images. Although the median edge detection works on pixel values ​​in the spatial domain, the present invention is used to predict alternative HH coefficients. In the present invention, the median edge detection predictor uses three adjacent coefficients of the actual HH coefficient for prediction, and the predicted value is represented by h. The prediction rules of the predictor are summarized as follows:

[0143]

[0144] Among them, a is the first coefficient value of the first adjacent coordinate point, b is the second coefficient value of the second adjacent coordinate point, and c is the third coefficient value of the third adjacent coordinate point. The crop image to be compressed is reconstructed by performing inverse wavelet transform using HH with prediction coefficients.

[0145] In step 204, all detail parameters are processed according to the inverse discrete wavelet transform IDWT to reconstruct the crop image to be compressed. In an optional embodiment, the present invention uses the peak signal-to-noise ratio (PSNR) and the root mean square difference percentage (PRD) of the reconstructed image quality to evaluate the performance of the agricultural multimodal data joint compression method. The peak signal-to-noise ratio is calculated as a function of the root mean square (MSE) as follows:

[0146]

[0147] In formula (5), I(i,j) is the original image with size N×M and B bits / pixel, is the reconstructed image consisting of the predicted values.

[0148]

[0149] Where I is the crop image to be compressed, is the reconstructed crop image to be compressed, B is the bit depth of the image, and the root mean square difference percentage PRD is calculated as follows:

[0150]

[0151] In formula (7), s k is the environmental time series data, It is the environmental timing data determined according to the modified scale factor and the replaced high-frequency detail parameter.

[0152] Optionally, according to the agricultural multimedia data joint compression method provided by the present invention, the receiving of the compressed bit stream from the codec and obtaining the replaced diagonal high-frequency detail parameters include:

[0153] generating a mixed crop image according to the compressed bit stream to generate mixed crop data according to the mixed crop image;

[0154] The replaced diagonal high-frequency detail parameters are determined according to the mixed crop data and all the detail parameters.

[0155] In such an embodiment, the mixed crop data is not decompressed, but a mixed crop image is first generated based on the compressed bit stream, and then mixed crop data is generated based on the mixed crop image. The process of generating mixed crop data based on the mixed crop image can refer to the process of processing the crop image to be compressed based on discrete wavelet transform DWT, and then after determining the mixed crop data, the replaced diagonal high-frequency detail parameters are determined based on the mixed crop data and all the detail parameters.

[0156] Those skilled in the art understand that the traditional agricultural Internet of Things transmits environmental parameters or crop phenotypic information by setting fixed time intervals, ignoring the correlation between the gradual change of the agricultural environment and heterogeneous data, which will lead to redundant data transmission and complicated data processing. The present invention proposes to transmit only the maximum value, minimum value and average value after removing the maximum value and minimum value of the environmental parameters related to temperature, humidity and light during the period according to the crop image data collection interval. After the multimedia sensor receives the environmental data during the period, it collects image data and then jointly compresses the crop image data and representative environmental data. The present invention can selectively transmit environmental scalar data, effectively reduce data redundancy, and reduce network energy consumption. And the associated transmission of environmental parameters and images can simplify the data spatiotemporal alignment and processing complexity.

[0157] It is worth noting that the growth status of crops and their related environmental changes have temporal and spatial correlation characteristics. Existing research on agricultural decision-making, crop inversion, etc. mainly aligns crop images with their corresponding environmental parameters through manual screening, and lacks automatic alignment of heterogeneous data based on the Internet of Things. In addition, the energy of wireless sensor network nodes is limited, and the energy consumption of high-throughput multimedia data and scalar data transmission is large, which will lead to premature death of nodes and network failure. The present invention is based on the characteristics of wavelet transform and wavelet image encoder, and only uses one codec to jointly compress image and environmental information. It can reduce computational complexity, thereby speeding up the encoding / decoding process; secondly, the multimodal compression scheme proposed in the present invention defines two new environmental scalar / crop image merging and separation stages in the encoding and decoding process, respectively, thereby producing efficient compression performance, which not only reduces storage space, but is also the key to ensuring fast and efficient data transmission.

[0158] The present invention provides a method, device, equipment and medium for joint compression of agricultural multimedia data. The present invention obtains all detail parameters of the crop image to be compressed in the wavelet domain and the wavelet transform coefficients of environmental time series data according to discrete wavelet transform DWT, determines diagonal high-frequency detail parameters from all detail parameters, selects target detail parameters corresponding to the parameter area to be replaced from the diagonal high-frequency detail parameters, inserts the wavelet transform coefficients of the environmental time series data into the parameter area to be replaced, determines the diagonal high-frequency detail parameters after replacement, determines mixed crop data according to the diagonal high-frequency detail parameters after replacement and all detail parameters, generates a compressed bit stream after compressing the mixed crop data, and sends the compressed bit stream to a decision center. The present invention uses the same encoder to compress and reconstruct agricultural spatiotemporal data of different standard formats, while maintaining good reconstruction quality and decision accuracy, it can minimize the overall data rate, reduce the energy consumption of agricultural hybrid wireless sensor networks, and extend the network life. The present invention not only reduces transmission energy consumption through efficient compression performance, but also can align multi-source data associated in time and space, simplify the data processing process, and help improve the accuracy of agricultural situation deduction, thereby achieving accurate and intelligent agricultural decision-making.

[0159] Figure 6 This is the third flow chart of the joint image processing method provided by the present invention, such as Figure 6As shown, in the left branch, the crop image to be compressed is quantized and decomposed by a 9 / 7 biorthogonal filter and a 2D wavelet to obtain the spatial domain corresponding to the high-frequency detail parameters. In the right branch, the environmental scalar data is quantized and decomposed by a 9 / 7 biorthogonal filter and a 1D wavelet to obtain the environmental time series data. According to the wavelet quantization signal of the i-th environmental parameter, the environmental time series data is inserted into the spatial domain corresponding to the high-frequency detail parameter. A mixed image is generated by a 2D wavelet inverse transform. The mixed image (wavelet coefficients) is encoded by an encoder. Finally, a suitable bit stream is used to compress the mixed image to obtain a compressed mixed crop image.

[0160] Figure 7 This is the fourth flow chart of the joint image processing method provided by the present invention. Figure 7 Disclosed is a joint image processing method that combines tasks performed by a codec, a sensor, and a decision center, wherein the crop phenotype image and environmental parameters are mixed according to a mixing function, and encoding is performed after the mixed data. After the decision center receives the coded compressed mixed crop image from the sensor, it decompresses the mixed image to obtain the mixed image, and then uses a separation function to separate the mixed data, determine the crop phenotype data prediction and the crop phenotype image reconstruction, and simultaneously realize the environmental parameter extraction and reconstruction of the environmental parameters.

[0161] Figure 8 It is one of the structural schematic diagrams of the joint image processing device provided by the present invention. The present invention discloses a joint image processing device, the execution subject is a codec, or a sensor with a codec, including an acquisition unit 11: used to process the crop image to be compressed and the environmental time series data respectively according to the discrete wavelet transform DWT, and obtain all detail parameters of the crop image to be compressed in the wavelet domain and the wavelet transform coefficients of the environmental time series data. The working principle of the acquisition unit 11 can refer to the aforementioned step 101 and will not be repeated here.

[0162] The joint image processing device also includes a processing unit 12: used to select target detail parameters corresponding to the parameter area to be replaced from the diagonal high-frequency detail parameters according to the length characteristics of the wavelet transform coefficients of the environmental time series data. The working principle of the processing unit 12 can refer to the aforementioned step 102 and will not be repeated here.

[0163] The joint image processing device also includes a sending unit 13: used to insert the wavelet transform coefficients of the environmental time series data into the parameter area to be replaced, determine the diagonal high-frequency detail parameters after replacement, and send a compressed bit stream to the decision center. The working principle of the sending unit 13 can refer to the aforementioned step 103 and will not be repeated here.

[0164] The diagonal high-frequency detail parameter is determined from all the detail parameters;

[0165] The mixed crop data is determined according to the replaced diagonal high-frequency detail parameters and all the detail parameters;

[0166] The compressed bit stream is generated after compressing the mixed crop data;

[0167] The environmental time series data is determined based on environmental scalar data collected by the multimedia device, and the collection time of the environmental scalar data includes the time of collecting the crop image to be compressed, multiple time points before collecting the crop image to be compressed, and multiple time points after collecting the crop image to be compressed.

[0168] The present invention provides a method, device, equipment and medium for joint compression of agricultural multimedia data. The present invention obtains all detail parameters of the crop image to be compressed in the wavelet domain and the wavelet transform coefficients of environmental time series data according to discrete wavelet transform DWT, determines diagonal high-frequency detail parameters from all detail parameters, selects target detail parameters corresponding to the parameter area to be replaced from the diagonal high-frequency detail parameters, inserts the wavelet transform coefficients of the environmental time series data into the parameter area to be replaced, determines the diagonal high-frequency detail parameters after replacement, determines mixed crop data according to the diagonal high-frequency detail parameters after replacement and all detail parameters, generates a compressed bit stream after compressing the mixed crop data, and sends the compressed bit stream to a decision center. The present invention uses the same encoder to compress and reconstruct agricultural spatiotemporal data of different standard formats, while maintaining good reconstruction quality and decision accuracy, it can minimize the overall data rate, reduce the energy consumption of agricultural hybrid wireless sensor networks, and extend the network life. The present invention not only reduces transmission energy consumption through efficient compression performance, but also can align multi-source data associated in time and space, simplify the data processing process, and help improve the accuracy of agricultural situation deduction, thereby achieving accurate and intelligent agricultural decision-making.

[0169] Fig. 9 This is the second structural schematic diagram of the joint image processing device provided by the present invention. The present invention discloses a joint image processing device, the execution subject is a decision center, including a receiving unit 21: used to receive a compressed bit stream from a codec, and obtain the diagonal high-frequency detail parameters after replacement. The working principle of the receiving unit 21 refers to the aforementioned step 201 and will not be repeated here.

[0170] The joint image processing device also includes an extraction unit 22: used to extract the wavelet parameters of the environmental time series data according to the replaced diagonal high-frequency detail parameters, and obtain the environmental time series data using the inverse discrete wavelet transform IDWT. The working principle of the extraction unit 22 refers to the aforementioned step 202 and will not be repeated here.

[0171] The joint image processing device further includes a determination unit 23: configured to determine the diagonal high-frequency detail parameter according to the replaced diagonal high-frequency detail parameter and the predicted value. The working principle of the determination unit 23 refers to the aforementioned step 203 and will not be described in detail here.

[0172] The joint image processing device further includes a reconstruction unit 24 for processing all detail parameters according to inverse discrete wavelet transform IDWT to reconstruct the crop image to be compressed. The working principle of the reconstruction unit 24 refers to the aforementioned step 204 and will not be described in detail here.

[0173] The predicted value is determined according to the adjacent coordinates of the coordinate values ​​of the rectangular area function of the diagonal high-frequency detail parameter after the replacement;

[0174] The replaced diagonal high-frequency detail parameters are determined according to the mixed crop data and all the detail parameters;

[0175] The mixed crop data is generated after decompressing the compressed bit stream.

[0176] The present invention provides a method, device, equipment and medium for joint compression of agricultural multimedia data. The present invention obtains all detail parameters of the crop image to be compressed in the wavelet domain and the wavelet transform coefficients of environmental time series data according to discrete wavelet transform DWT, determines diagonal high-frequency detail parameters from all detail parameters, selects target detail parameters corresponding to the parameter area to be replaced from the diagonal high-frequency detail parameters, inserts the wavelet transform coefficients of the environmental time series data into the parameter area to be replaced, determines the diagonal high-frequency detail parameters after replacement, determines mixed crop data according to the diagonal high-frequency detail parameters after replacement and all detail parameters, generates a compressed bit stream after compressing the mixed crop data, and sends the compressed bit stream to a decision center. The present invention uses the same encoder to compress and reconstruct agricultural spatiotemporal data of different standard formats, while maintaining good reconstruction quality and decision accuracy, it can minimize the overall data rate, reduce the energy consumption of agricultural hybrid wireless sensor networks, and extend the network life. The present invention not only reduces transmission energy consumption through efficient compression performance, but also can align multi-source data associated in time and space, simplify the data processing process, and help improve the accuracy of agricultural situation deduction, thereby achieving accurate and intelligent agricultural decision-making.

[0177] Fig.10 Schematic diagram of the structure of the electronic device provided by the present invention. Fig.10As shown, the electronic device may include: a processor (processor) 110, a communication interface (Communications Interface) 120, a memory (memory) 130 and a communication bus 140, wherein the processor 110, the communication interface 120, and the memory 130 communicate with each other through the communication bus 140. The processor 110 can call the logic instructions in the memory 130 to execute the agricultural multimedia data joint compression method, which includes: processing the crop image to be compressed and the environmental time series data respectively according to the discrete wavelet transform DWT, obtaining all detail parameters of the crop image to be compressed in the wavelet domain and the wavelet transform coefficients of the environmental time series data; selecting the target detail parameters corresponding to the parameter area to be replaced from the diagonal high-frequency detail parameters according to the length characteristics of the wavelet transform coefficients of the environmental time series data; inserting the wavelet transform coefficients of the environmental time series data into the parameter area to be replaced, determining the diagonal high-frequency detail parameters after replacement, and sending a compressed bit stream to the decision center; the diagonal high-frequency detail parameters are determined from all the detail parameters; the mixed crop data is determined according to the diagonal high-frequency detail parameters after replacement and all the detail parameters; the compressed bit stream is generated after compressing the mixed crop data; the environmental time series data is determined according to the environmental scalar data collected by the multimedia device, and the collection time of the environmental scalar data includes the time of collecting the crop image to be compressed, multiple time points before collecting the crop image to be compressed, and multiple time points after collecting the crop image to be compressed.

[0178] In addition, the logic instructions in the above-mentioned memory 130 can be implemented in the form of software functional units and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program code.

[0179] On the other hand, the present invention also provides a computer program product, which includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute a method for joint compression of agricultural multimedia data provided by the above methods. The method includes: processing the crop image to be compressed and the environmental time series data respectively according to the discrete wavelet transform DWT to obtain all detail parameters of the crop image to be compressed in the wavelet domain and the wavelet transform coefficients of the environmental time series data; selecting the target details corresponding to the parameter area to be replaced from the diagonal high-frequency detail parameters according to the length characteristics of the wavelet transform coefficients of the environmental time series data. Parameters; inserting the wavelet transform coefficients of the environmental time series data into the parameter area to be replaced, determining the diagonal high-frequency detail parameters after replacement, and sending a compressed bit stream to the decision center; the diagonal high-frequency detail parameters are determined from all the detail parameters; the mixed crop data is determined based on the diagonal high-frequency detail parameters after replacement and all the detail parameters; the compressed bit stream is generated after compressing the mixed crop data; the environmental time series data is determined based on the environmental scalar data collected by the multimedia device, and the collection time of the environmental scalar data includes the time of collecting the crop image to be compressed, multiple times before collecting the crop image to be compressed, and multiple times after collecting the crop image to be compressed.

[0180] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which is implemented when the processor executes the agricultural multimedia data joint compression method provided by the above methods, the method comprising: processing the crop image to be compressed and the environmental time series data respectively according to the discrete wavelet transform DWT, obtaining all detail parameters of the crop image to be compressed in the wavelet domain and the wavelet transform coefficients of the environmental time series data; selecting the target detail parameters corresponding to the parameter area to be replaced from the diagonal high-frequency detail parameters according to the length characteristics of the wavelet transform coefficients of the environmental time series data; inserting the wavelet transform coefficients of the environmental time series data into the target detail parameters corresponding to the diagonal high-frequency detail parameters; and The coefficients are converted to the parameter area to be replaced, the diagonal high-frequency detail parameters after replacement are determined, and a compressed bit stream is sent to a decision center; the diagonal high-frequency detail parameters are determined from all the detail parameters; the mixed crop data is determined based on the diagonal high-frequency detail parameters after replacement and all the detail parameters; the compressed bit stream is generated after compressing the mixed crop data; the environmental time series data is determined based on the environmental scalar data collected by the multimedia device, and the collection time of the environmental scalar data includes the time of collecting the crop image to be compressed, multiple time points before collecting the crop image to be compressed, and multiple time points after collecting the crop image to be compressed.

[0181] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0182] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0183] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for joint compression of agricultural multimedia data, characterized in that: include: Process the crop image to be compressed and the environmental time series data respectively according to discrete wavelet transform DWT to obtain all detail parameters of the crop image to be compressed in the wavelet domain and the wavelet transform coefficients of the environmental time series data; Selecting target detail parameters corresponding to the parameter area to be replaced from the diagonal high-frequency detail parameters according to the length characteristics of the wavelet transform coefficients of the environmental time series data; the said all detail parameters include the said diagonal high-frequency detail parameters and other detail parameters; Inserting the wavelet transform coefficients of the environmental time series data into the parameter area to be replaced, determining the high-frequency detail parameters in the diagonal direction after the replacement, and sending a compressed bit stream to the decision center; The step of determining the high-frequency detail parameters of the diagonal direction after replacement and sending the compressed bit stream to the decision center comprises: Determine mixed crop data according to the replaced diagonal high-frequency detail parameter and other detail parameters among all the detail parameters; Restoring a mixed crop image according to the mixed crop data, compressing the mixed crop image to generate the compressed bit stream, and sending the compressed bit stream to the decision center; the process of restoring the mixed crop image according to the mixed crop data is the inverse process of processing the crop image to be compressed according to discrete wavelet transform DWT; The diagonal high-frequency detail parameter is determined from all the detail parameters; The compressed bit stream is generated after compressing the mixed crop data; The environmental time series data is determined based on environmental scalar data collected by the multimedia device, and the collection time of the environmental scalar data includes the time of collecting the crop image to be compressed, multiple time points before collecting the crop image to be compressed, and multiple time points after collecting the crop image to be compressed.

2. The agricultural multimedia data joint compression method according to claim 1, characterized in that: The step of selecting target detail parameters corresponding to the parameter area to be replaced from the diagonal high-frequency detail parameters according to the length characteristics of the wavelet transform coefficients of the environmental time series data includes: Constructing a spatial domain corresponding to the diagonal high-frequency detail parameter according to the diagonal high-frequency detail parameter; The target detail parameters are determined from the spatial domain according to the wavelet transform coefficients, and the region where the target detail parameters are located is determined as the parameter region to be replaced.

3. The agricultural multimedia data joint compression method according to claim 1, characterized in that: The step of inserting the wavelet transform coefficients of the environmental time series data into the parameter region to be replaced and determining the diagonal high-frequency detail parameters after replacement includes: Determine the maximum value of the wavelet transform coefficients in all environmental time series data, and determine the maximum value as the correction scale factor; The replaced diagonal high-frequency detail parameter is determined according to the correction scale factor and the parameter area to be replaced.

4. The agricultural multimedia data joint compression method according to claim 1, characterized in that: The method of processing the crop image to be compressed and the environmental time series data respectively according to the discrete wavelet transform DWT to obtain all detail parameters of the crop image to be compressed in the wavelet domain and the wavelet transform coefficients of the environmental time series data includes: Using two-dimensional discrete wavelet transform DWT to process the crop image to be compressed, and obtaining all detail parameters of the crop image to be compressed in the wavelet domain, wherein all detail parameters at least include the diagonal high-frequency detail parameters; The environmental time series data is processed using one-dimensional discrete wavelet transform (DWT) to obtain the wavelet transform coefficients of the environmental time series data.

5. A method for joint compression of agricultural multimedia data, characterized in that: include: receiving a compressed bit stream from a codec, obtaining mixed crop data from the mixed crop image, and obtaining a replaced diagonal high-frequency detail parameter; Extracting wavelet parameters of environmental time series data according to the replaced diagonal high-frequency detail parameters, and acquiring environmental time series data by using inverse discrete wavelet transform IDWT; Determine the diagonal high-frequency detail parameter according to the replaced diagonal high-frequency detail parameter and the predicted value; Processing the diagonal high-frequency detail parameters and other detail parameters among all detail parameters according to the inverse discrete wavelet transform IDWT, reconstructing the crop image to be compressed; The receiving of the compressed bit stream from the codec and obtaining the replaced diagonal high frequency detail parameters comprises: Processing the mixed crop image according to discrete wavelet transform DWT to obtain mixed crop data; All the detail parameters of the mixed crop data include other detail parameters and replaced diagonal high-frequency detail parameters; The predicted value is determined according to the adjacent coordinates of the coordinate values ​​of the rectangular area function of the diagonal high-frequency detail parameter after the replacement.

6. The agricultural multimedia data joint compression method according to claim 5, characterized in that: The step of extracting the wavelet parameters of the environmental time series data according to the replaced diagonal high-frequency detail parameters and acquiring the environmental time series data by using inverse discrete wavelet transform (IDWT) comprises: Determine the wavelet parameters of the environmental time series data according to the quotient of the diagonal high-frequency detail parameter after the replacement and the correction scale factor; The wavelet parameters for acquiring the environmental time series data are processed by using one-dimensional inverse discrete wavelet transform (IDWT) to determine the environmental time series data.

7. The agricultural multimedia data joint compression method according to claim 5, characterized in that: The determining the diagonal high-frequency detail parameter according to the replaced diagonal high-frequency detail parameter and the predicted value comprises: Determine a first coefficient value of a first adjacent coordinate point, a second coefficient value of a second adjacent coordinate point, and a third coefficient value of a third adjacent coordinate point according to the coordinate value of the rectangular area function in the space domain corresponding to the diagonal high-frequency detail parameter after the replacement; When the maximum value of the first coefficient value or the second coefficient value is less than the third coefficient value, the predicted value is the minimum value of the first coefficient value or the second coefficient value; In the case where the minimum value of the first coefficient value or the second coefficient value is greater than or equal to the third coefficient value, the predicted value is the maximum value of the first coefficient value or the second coefficient value; In the case where the maximum value of the first coefficient value or the second coefficient value is greater than or equal to the third coefficient value, or in the case where the minimum value of the first coefficient value or the second coefficient value is less than the third coefficient value, the predicted value is determined based on the sum of the first coefficient value and the second coefficient value and based on the difference between the sum and the third coefficient value; Among them, the first adjacent coordinate point is the left pixel coordinate value of the rectangular area function coordinate value, the second adjacent coordinate point is the upper pixel coordinate value of the rectangular area function coordinate value, and the third adjacent coordinate point is the upper left pixel coordinate value of the rectangular area function coordinate value.

8. An agricultural multimedia data joint compression device, characterized in that: include: An acquisition unit is used to process the crop image to be compressed and the environmental time series data respectively according to discrete wavelet transform DWT, and acquire all detail parameters of the crop image to be compressed in the wavelet domain and the wavelet transform coefficients of the environmental time series data; A processing unit: used for selecting target detail parameters corresponding to the parameter area to be replaced from the diagonal high-frequency detail parameters according to the length characteristics of the wavelet transform coefficients of the environmental time series data; the said all detail parameters include the said diagonal high-frequency detail parameters and other detail parameters; A sending unit: used for inserting the wavelet transform coefficients of the environmental time series data into the parameter area to be replaced, determining the high-frequency detail parameters in the diagonal direction after replacement, and sending a compressed bit stream to the decision center; The step of determining the high-frequency detail parameters of the diagonal direction after replacement and sending the compressed bit stream to the decision center comprises: Determine mixed crop data according to the replaced diagonal high-frequency detail parameter and other detail parameters among all the detail parameters; Restoring a mixed crop image according to the mixed crop data, compressing the mixed crop image to generate the compressed bit stream, and sending the compressed bit stream to the decision center; the process of restoring the mixed crop image according to the mixed crop data is the inverse process of processing the crop image to be compressed according to discrete wavelet transform DWT; The diagonal high-frequency detail parameter is determined from all the detail parameters; The compressed bit stream is generated after compressing the mixed crop data; The environmental time series data is determined based on environmental scalar data collected by the multimedia device, and the collection time of the environmental scalar data includes the time of collecting the crop image to be compressed, multiple time points before collecting the crop image to be compressed, and multiple time points after collecting the crop image to be compressed.

9. An agricultural multimedia data joint compression device, characterized in that: include: Receiving unit: used for receiving the compressed bit stream from the codec, and obtaining the replaced diagonal high-frequency detail parameters; An extraction unit: used for extracting the wavelet parameters of the environment time series data according to the replaced diagonal high-frequency detail parameters, and obtaining the environment time series data by using inverse discrete wavelet transform IDWT; A determining unit: configured to determine the diagonal high-frequency detail parameter according to the replaced diagonal high-frequency detail parameter and the predicted value; Reconstruction unit: used for processing the diagonal high-frequency detail parameter and other detail parameters in all detail parameters according to the inverse discrete wavelet transform IDWT, so as to reconstruct the crop image to be compressed; The receiving of the compressed bit stream from the codec and obtaining the replaced diagonal high frequency detail parameters comprises: Processing the mixed crop image according to discrete wavelet transform DWT to obtain mixed crop data; All the detail parameters of the mixed crop data include other detail parameters and replaced diagonal high-frequency detail parameters; The predicted value is determined according to the adjacent coordinates of the coordinate values ​​of the rectangular area function of the diagonal high-frequency detail parameter after the replacement.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the agricultural multimedia data joint compression method as described in any one of claims 1-7 is implemented.

11. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the agricultural multimedia data joint compression method as described in any one of claims 1 to 7 is implemented.

Images