CN103377189A - Method for collecting ground reference information of grain on basis of intelligent mobile terminals - Google Patents

Abstract

The invention discloses a method for collecting ground reference information of grain on the basis of intelligent mobile terminals. The method is characterized in that the ground reference information is collected during grain production on the basis of the existing universal intelligent mobile phone terminals, and the method is used for analyzing remote sensing data. The method has the advantages that a data mining technology is utilized, the correctness of acquisition results is evaluated by the aid of systematic analysis for user feedback results, a corresponding user information collecting database is created, and contents which can be skillfully collected by users can be described, so that corresponding tasks can be recommended, and the accuracy of the acquired information is improved.

Description

Based on the grain ground of intelligent mobile terminal with reference to formation gathering method

One, technical field

The present invention is based on now widely used mobile hand-held device operating system iOS or Android development environment, take full advantage of the powerful functions such as embedded type sensor in the 3G smart mobile phones such as present iphone, htc, comprise high-definition camera, the high-precision GPS positioning system, infrared thermography, a series of embedded devices such as photosensitive sensor, acceleration transducer, the relevant grain ground reference of dynamic acquisition information, and send in real time remote server by wireless communication technology.

Two, background technology

For this open physical environment of extensive Farming, environmental factor comprises that temperature, humidity, illumination, height above sea level, soil parameters (soil types, soil moisture, P in soil H value etc.) etc. can not be ignored the impact in plant growth cycle.And in recent years, fast development along with mobile communication technology, remote sensing technology, computer technology, embedded type sensor technology etc.,, enriching the range of application of technology how with these technological incorporation together, is one of numerous scientific workers Major Strategic problem of demanding urgently studying.Based on this, we utilize the GPS positioning function in the smart mobile phone in the present invention, and the spatial informations such as the sea level elevation of the current Farming of automatic acquisition and longitude and latitude, and coupling Satellite Remote Sensing are corrected in real time; Gather simultaneously other physical parameters, and its error is classified, proofreaied and correct, thereby then the accuracy that improves information is merged with other relevant informations and be committed to server terminal.

Three, summary of the invention

The equipment of this invention by match user uses ability and with reference to the complexity of information task, automatically being fit to of task is recommended the user, consider simultaneously user's the comprehensive key elements such as commuting cost, time cost, input cost, thus the cost that reduction information is used.

Grainfield terrestrial information collection method based on general purpose transducer, the method is based on the relevant built-in sensors in the intelligent mobile terminal, measure the environmental parameter in the Farming, and carry out error correction by the method that software is realized, thereby improve the accuracy of information.

Information is uploaded coupling interlock system in real time.The via satellite analysis of orbit so that terrestrial information acquisition time and remotely-sensed data acquisition time basic synchronization, thereby so that the ground that obtains is carried out Auto-matching with reference to information according to value and capacity and satellite remote sensing date, makes up the interlock system.Simultaneously, terrestrial information and the satellite data of obtaining in time can be transferred to control end, to realize the Real Time Monitoring to Farming.

Embodiment of the present invention provides a kind of user task proposed algorithm based on collaborative filtering (CF), and its particular content is as follows:

Collaborative filtering is called again collaborative filtering or social filtering, and it is based on following hypothesis: if some users are more similar to the scoring of some projects, then they are also more similar to the scoring of other project.So the basic thought of its realization is that system adopts certain technology to find some neighbours (user of similar interests being arranged with the targeted customer) of targeted customer, then according to the scoring of nearest-neighbors to destination item, produces and recommends.The collaborative filtering analysis be the relation of " user one user ", recommend resource according to the similarity of user interest.

Method based on user's collaborative filtering algorithm vacuum metrics user similarity mainly contains following two kinds: cosine similarity and relevant similarity, computing formula are respectively suc as formula shown in (1) and the formula (2):

$\mathrm{Sim}(u,n)=\frac{\underset{m=1}{\overset{N}{\mathrm{\Σ}}}{R}_{u,m}*R_{n,m},}{\sqrt{\underset{m=1}{\overset{N}{\mathrm{\Σ}}}{R}_{u,m}*\underset{m=1}{\overset{N}{\mathrm{\Σ}}}{R}_{n,m}}}---\left(1\right)$

$\mathrm{Sim}(u,n)=\frac{\underset{c\∈I}{\mathrm{\Σ}}(R_{u,c}-{\stackrel{\‾}{R}}_u)*(R_{n,c}-{\stackrel{\‾}{R}}_n)}{\sqrt{\underset{c\∈I}{\mathrm{\Σ}}{(R_{u,c}-{\stackrel{\‾}{R}}_u)}^2*\underset{c\∈I}{\mathrm{\Σ}}{(R_{n,c}-{\stackrel{\‾}{R}}_n)}^2}}---\left(2\right)$

In the formula: N---the number of resource;

R _{U, m}And R _{N, m}---user u and n represent the scoring to resource item m;

I---the Item Sets that user u and n marked jointly;

R _{U, c}And R _{N, c}---user u and n are to the scoring of a certain resource among the I;

With

---user u and n are to the average score of resource.

During information recommendation, according to nearest-neighbors the scoring of project u is produced and to recommend, adopts following formula (3) calculating targeted customer u that the prediction of scoring item i is not marked:

$\mathrm{Pred}(u,i)={\stackrel{\‾}{R}}_u+\frac{\underset{a=1}{\overset{b}{\mathrm{\Σ}}}(R_{a,i}-{\stackrel{\‾}{R}}_a)*\mathrm{Sim}(u,a)}{\underset{a=1}{\overset{b}{\mathrm{\Σ}}}\mathrm{Sim}(u,a)}---\left(3\right)$

In the formula:

---user u and its arest neighbors a are to the average score of resource item, R _{A, i}---arest neighbors a is to the scoring of project i, the number of b---arest neighbors, Sim (u, a)---the similarity of user u and a.

User u to the adaptedness of task i is: the user is to the average fitness of all tasks

Add the customer group the most similar to it

Poor to the adaptedness of task i adaptedness average with it.By to this formula, just can determine those tasks are recommended corresponding user.

The present invention proposes a kind of based on user collaboration filtering information recommended models, this model is on the basis to the resource item classification, scoring is converted to the user to the scoring of resource item classification to project with the user, after the logarithm Data preprocess, to user clustering, at last the targeted customer place bunch in seek its nearest-neighbors and produce and recommend.Simultaneously, can be by extendability and the recommendation quality of weighting filtering data pre-service to improve algorithm.

Four, description of drawings

Fig. 1 Terminal Server Client function structure

Fig. 2 Farming terrestrial information collecting flowchart figure

Five, embodiment

The method that embodiment of the present invention provides GPS locating information acquisition and farmland general information collection to combine, and remote sensing technology via satellite, the accuracy of coupling institute Information Monitoring, and proofreaied and correct and data fusion, send to server terminal by wireless transmission at last, concrete steps are as follows:

Step 1 start-up system is carried out self check, initialize routine;

Step 2 starts GPS and intelligent field information acquisition sensor, and GPS gathers relevant height above sea level, latitude and longitude information, and whether acquisition system inquiry in farmland opens high-definition camera, in this way, then gather image information, otherwise, gather temperature, humidity, the conditional informations such as soil, illumination are also preserved;

The data that the terrestrial information that step 3 gathers by GPS and satellite remote sensing provide compare, and if any error, then revise the GPS information parameter, then are saved in the storage card;

Step 4 sends to remote server terminal by cordless communication network after the data of preserving are gathered, and real-time dynamic display;

Step 5 this moment, the farmland dynamic implantation information that collects all can embody user's resource requirement, and according to certain principle, as by the weighted mean filtering data etc., the overall situation draws dynamic real-time curve, and returns.

Claims (4)

1. One kind based on the grain ground of intelligent mobile terminal with reference to formation gathering method, the feature of its method implementation procedure is specific as follows: utilize universal intelligent portable terminal related sensor, design grain ground is with reference to formation gathering method; By the user being uploaded result's evaluation and excavation, analysis user in collection process characteristics and set up corresponding database; , carry out task and recommend with reference to the information feature based on user's ground, realize the Data Collection of efficiently and accurately.
2. According to the described grain ground based on intelligent mobile terminal of claims 1 with reference to formation gathering method, its Data Collection is characterised in that: the sensor that takes full advantage of the General Mobile intelligent terminal, design corresponding ground with reference to formation gathering method, to reduce the cost of Data Collection.Particularly, at first need to analyse in depth the sensor that existing mobile device such as smart mobile phone comprise, such as camera, thermometer, GPS orientator, gyroscope etc.Take full advantage of above intelligence sensor by software mode, when not increasing hardware cost, obtain the ground comparable data that satisfies accuracy requirement.
3. 3. upload result's evaluation and method for digging according to claims 1 described user, it is characterized in that by utilizing collaborative filtering, by the cluster to the user, the excavation similar users is also uploaded information with this to it and is carried out cross validation, thereby ensures the accuracy of Data Collection.
4. 4. according to claims 1 described user information collection database, it is characterized in that this database is automatically to create and renewal by collaborative filtering, namely by the comparison to the user profile of obtaining and the ground comparable data of uploading thereof, automatically set up the user and upload historical data base, and carry out according to this information commending contents, obtain simultaneously feedback information, more new database.

Images