JP3984135B2 - Moving locus data search information generation apparatus and method, movement locus data search apparatus and method, movement locus data search information generation program and recording medium recording the program, movement locus data search program and program thereof Recorded recording medium - Google Patents

Description

という５つの関係を問い合わせに用いれば足りる。
【０１１１】
次に、これらの問い合わせに対する答えを検索する方法について説明する。
【０１１２】
（Ｉ）Ｅnter：
問い合わせ式がＱ((Ｒ，Ｅenter ))の場合、先ず最初に、Ｒを通る移動軌跡データをインデックスを使って探す。次に、各移動軌跡データのリンクリストを使って、Ｒの直前に別の分割矩形Ｒ' を通っているかどうかを調べる（Ｒがリンクリストの先頭でない限りは、Ｒの直前にＲの四方のいずれかの分割矩形を通っている）。Ｒ' が存在する場合は、その移動軌跡データを答えとして返す。
【０１１３】
（II）Ｅxit ：
問い合わせ式がＱ((Ｒ，Ｅexit))の場合、先ず最初に、Ｒを通る移動軌跡データをインデックスを使って探す。次に、各移動軌跡データのリンクリストを使って、Ｒの直後に別の分割矩形Ｒ' を通っているかどうかを調べる（Ｒがリンクリストの末尾でない限りは、Ｒの直後にＲの四方のいずれかの分割矩形を通っている）。Ｒ' が存在する場合は、その移動軌跡データを答えとして返す、
（III)Ｒeturn ：
問い合わせ式がＱ((Ｒ，Ｅreturn))の場合、先ず最初に、Ｒを通る移動軌跡データをインデックスを使って探す。次に、各移動軌跡データのリンクリストを使って、Ｒから順番にリンクをたどり、またＲに戻るようなルートがリンクリスト上にあるかどうかを調べる。もしそのようなルートが存在すれば、その移動軌跡データを答えとして返す。
【０１１４】
（IV) Ｃross：
問い合わせ式がＱ((Ｒ，Ｅcross ))の場合、先ず最初に、ＲについてＥenter の条件を調べ、その後、さらにＥexitのケースについて調べる。両方を満たす移動軌跡データを答えとして返す。
【０１１５】
（Ｖ）Ｐath ：
問い合わせ式がＱ((Ｒ₁,Ｒ₂,Ｅpath))の場合、先ず最初に、Ｒ₁ を通る移動軌跡データをインデックスを使って探す。次に、各移動軌跡データのリンクリストを使って、Ｒ₁ から順番にリンクをたどりＲ₂ にたどり着くようなルートがあるかどうかを調べる。もしそのルートが存在すれば、その移動軌跡データを答えとして返す。
【０１１６】
以上の処理によつて得られた答えの中から、条件式に含まれている時刻に関する条件を調べる、検索部１２は、最終的に時刻条件も満たす移動軌跡データだけを答えとして返す。
【０１１７】
次に、検索式で指定される領域Ｒがインデックスのために作られた各分割矩形と一致しない場合の検索方法について述べる。
【０１１８】
（Ａ）検索式で指定されるＲがτ₀ よりも大きい（インデックスが対象としている範囲を超えている）場合には、インデックスを使用することができないので、移動軌跡データすべてについて与えられた条件を直接調べることで、検索式を充足する移動軌跡データを検索する。
【０１１９】
（Ｂ）検索式で指定されるＲがインデックスの分割矩形階層の最下層の分割矩形のいずれかに完全に含まれる場合には、次のように計算する。
【０１２０】
（Ｂ−１）Ｅnter／Ｅxit ／Ｒeturn ／Ｃrossの場合：
Ｒが含まれる最下層の分割矩形をＲ' とすると、先ず最初に、インデックスを使って、Ｒ' を通過する移動軌跡データを探す。その後に、リンクリストを使わず、それぞれの移動軌跡データについて条件のチェックを行う。
【０１２１】
（Ｂ−２）Ｐath の場合：
Ｒ₁ が含まれる最下層の分割矩形をＲ₁'、Ｒ₂ が含まれる最下層の分割矩形をＲ₂'とすると、先ず最初に、インデックスを使って、Ｒ₁'を通った後にＲ₂'を通る移動軌跡データを調べ、その後に、リンクリストを使わず、Ｑ((Ｒ₁,Ｒ₂,Ｅ_path))を満たすかどうかを調べる。
【０１２２】
（Ｃ）検索式で指定されるＲがインデックス分割矩形階層の最下層の分割矩形のいずれにも完全には含まれない場合には、Ｒを含むインデックス中の最小の分割矩形を使って、リンクリストを調べた後、インデックスを使わずに、検索式を充足する移動軌跡データを検索する。その調べ方については、Ｒが最下層の分割矩形に完全に含まれる場合と同様である。
【０１２３】
リンクリストは、特にＰath を含むクエリを計算する上で有効である。例えば、Ｒ₁, ... ,Ｒ_n を順番に通過していくような多数の分割矩形を含むＰath クエリでは、リンクリストをたどり、移動軌跡がＲ₁ からＲ_n まで順番に通過しているかどうかを簡単に調べることができる。
【０１２４】
図１７ないし図２１に、以上に説明した移動軌跡データ検索処理の処理フローを図示する。
【０１２５】
ここで、図１７及び図１８の処理フローが骨格となる処理フローであり、図１９及び図２０の処理フローは、図１７及び図１８の処理フローのステップ２３／ステップ３０で実行する移動軌跡データの検索処理の詳細な処理を示し、図２１の処理フローは、図１７及び図１８の処理フローのステップ３４で実行する移動軌跡データの検索処理の詳細な処理を示す。
【０１２６】
図１７及び図１８の処理フローについて簡単に説明するならば、ステップ２１，２２で、インデックスの最上位層の分割矩形（分割元となった矩形）と検索式で指定される領域Ｒの大きさを比較して、一致する場合には、図１９及び図２０の処理フローに従って、上述した（Ｂ）と同様の手順に従い、インデックス（リンクリストを含む）を使って、検索式を充足する移動軌跡データを検索する。
【０１２７】
そして、インデックスの最上位層の分割矩形（分割元となった矩形）と検索式で指定される領域Ｒの大きさとが一致しない場合には、ステップ２５で、検索式で指定される領域Ｒの方が大きいのか否かを判断して、検索式で指定される領域Ｒの方が大きい場合には、インデックスを用いることができないので、上述した（Ａ）で説明したように、インデックスを使わずに、検索式を充足する移動軌跡データを検索する。
【０１２８】
一方、ステップ２５で、検索式で指定される領域Ｒの方が小さい場合には、ステップ２８〜ステップ３３の処理を繰り返していくことで、インデックスの１つ下位層の分割矩形と検索式で指定される領域Ｒの大きさとが一致するのか否かを判断して、一致する場合には、図１９及び図２０の処理フローに従って、上述した（Ｂ）と同様の手順に従い、インデックス（リンクリストを含む）を使って、検索式を充足する移動軌跡データを検索する。
【０１２９】
そして、ステップ２８〜ステップ３３の処理を繰り返していくことで、最終的に、検索式で指定される領域Ｒの大きさとインデックスの分割矩形とが一致しないことを判断するときには、検索式で指定される領域Ｒを含む最小の分割矩形Ｒ' を特定して、ステップ３４で、図２１の処理フローに従って、その分割矩形Ｒ' が検索式で指定されたものとして扱って、インデックスを使って検索を行い、ステップ３５で、その検索した移動軌跡データの中から、インデックスを使わずに検索式を充足するものを検索する。
【０１３０】
すなわち、上述した（Ｃ）と同様の手順に従い、検索式で指定される領域Ｒを含む最小の分割矩形Ｒ' を特定すると、その分割矩形Ｒ' が検索式で指定されたものとして扱って、インデックスを使って検索を行い、その後、その検索した移動軌跡データの中から、インデックスを使わずに検索式を充足するものを検索するのである。
【０１３１】
ここで、図２１の処理フローで、Ｅnter／Ｅxit ／Ｒeturn ／Ｃrossに分けて検索を行わないのは、最小の分割矩形Ｒ' についてそのような条件を充足したとしても、検索式で指定される領域Ｒについてそのような条件を充足しているとは限らないからである。
【０１３２】
なお、いずれの検索ルートを通る場合にも、最終的な検索結果は、検索式で指定される時刻条件を充足するものを検索することで得ることになる。
【０１３３】
以上説明した実施形態例では、移動軌跡データとして２次元平面上のデータを扱っているが、本発明では、３次元空間上あるいはそれ以上の多次元空間内の移動軌跡データに対してもそのまま適用できるものである。この場合、移動軌跡データの各ポイントの座標を表すデータの個数がｎ次元空間内のデータである場合にはｎ個になる。
【０１３４】
また、本発明の取り扱う移動軌跡データは、移動する物体の移動軌跡データに限られるものではなく、複数のパラメータが連動して時間軸に沿って変化するような移動軌跡データに対してもそのまま適用できるものである。例えば、学生の身長、体重、胸囲、座高の４パラメータを各年に測定したケースでは、これらの４パラメータを軸とする四次元空間内の移動軌跡データとして扱うことができる。
【０１３５】
【発明の効果】
空間上を移動する物体の移動軌跡などを、物体の識別子と時刻と空間上の座標（点データ）との組として記録したものを移動軌跡データと呼ぶ。
【０１３６】
移動軌跡データは、一般には連続でない時系列に沿った離散的な点データ列である。移動軌跡データに含まれる点データ列を点データの時刻順に結んだものは、空間上では有限の長さを持ち向きがある線分になる。このような移動軌跡データを蓄積し検索する機能を有するデータベースシステムを、移動軌跡データベースシステムと呼ぶ。
【０１３７】
本発明では、このような移動軌跡データベースシステムに対して、空間上の複数の領域と移動軌跡データとの間の位相幾何学的な関係を検索条件として指定すると、移動軌跡データベースシステムが予め蓄積している移動軌跡データの通過するグリッドの順序を検出して作成するグリッドリンクと呼ぶインデックスを利用することで、指定された検索条件を満たす移動軌跡データを高速に検索することを実現する。
【図面の簡単な説明】
【図１】本発明を具備するデータベースシステムの全体構成図である。
【図２】移動軌跡データの一例を示す図である。
【図３】閉領域間の位相幾何学的関係を記述する行列式の説明図である。
【図４】閉領域と有向線分との間の位相幾何学的関係を記述する行列式の説明図である。
【図５】閉領域と有向線分との関係の説明図である。
【図６】閉領域と有向線分との間の位相幾何学的関係を記述する行列式の説明図である。
【図７】ＣＯＮＮＥＣＴ演算の説明図である。
【図８】ＣＯＮＮＥＣＴ演算の説明図である。
【図９】基本となる関係行列の説明図である。
【図１０】基本となる関係行列とＣＯＮＮＥＣＴ演算とにより導出される関係行列の説明図である。
【図１１】インデックスの説明図である。
【図１２】リンクリストの説明図である。
【図１３】矩形分割処理の説明図である。
【図１４】データ管理部の実行する処理フローである。
【図１５】データ管理部の実行する処理フローである。
【図１６】実用上で基本となる閉領域と有向線分との関係の説明図である。
【図１７】検索部の実行する処理フローである。
【図１８】検索部の実行する処理フローである。
【図１９】検索部の実行する処理フローである。
【図２０】検索部の実行する処理フローである。
【図２１】検索部の実行する処理フローである。
【符号の説明】
１ データベースシステム
１０ 入力部
１１ データ管理部
１２ 検索部
１３ 蓄積部[0001]
BACKGROUND OF THE INVENTION
  The present invention is used to search for movement trajectory data stored in a database.InformationSearch for moving trajectory data to be generatedInformation studentsGeneration apparatus and method, andSearch information forAnd a method and a method of searching for a moving track data that searches the moving track data stored in the database to satisfy a search condition, and the method of searching the moving track dataInformation studentsOf trajectory data used for the realization of an image forming deviceInformation studentsThe present invention relates to a composition program, a recording medium on which the program is recorded, a movement trajectory data search program used for realizing the movement trajectory data search apparatus, and a recording medium on which the program is recorded.
[0002]
It is required to search for movement trajectory data of a moving object in various systems such as a navigation system, a system for managing the position of a person, and a system for reading preference information from a movement trajectory of a person. In addition, various systems are required to search for data whose value changes along the time axis (moving trajectory data in a broad sense).
[0003]
From now on, it is necessary to construct a technique for realizing such search for movement trajectory data.
[0004]
[Prior art]
As prior arts related to the present invention, there are a time series database system, a spatio-temporal database system, and a navigation system.
[0005]
[1] Time series database system (for example, refer to Patent Document 1)
It may change along the time axis, such as economic index values similar to stock prices and exchange rates, measured values related to climate such as precipitation, temperature, and humidity, social index values similar to population, employment rate, and fertility rate Data obtained by measuring values at regular intervals or irregularly is generally referred to as time series data.
[0006]
The development of techniques for discovering periodicity and regularity from time-series data and predicting changes in future values has been carried out. In this technical development, a time series database system having a function of accumulating time series data and searching for time series data matching a condition has been developed.
[0007]
A conventional time-series database system has a function of searching for data that matches a search condition by giving a characteristic of increase / decrease change in the value of the data with time change as a search condition.
[0008]
The movement trajectory data describing the moving user is a kind of time-series data because it records the coordinate information of the user corresponding to each time.
[0009]
However, since the movement trajectory data has a property as a line segment in space, the search capability is not sufficient only by the search method that specifies the increase / decrease change of the value with respect to the time axis. It is necessary to provide a function for obtaining a data that matches a change in the shape of a geometric line segment and a change in a relationship with another figure around the line segment as conditions.
[0010]
However, existing time series database systems do not have this function.
[0011]
[2] Spatio-temporal database system (for example, see Patent Document 2)
There are a spatial database system and a spatio-temporal database system as database systems having a function capable of searching a figure in space by giving a relation between figures and a figure shape as conditions. These database systems are also used as a basic technology of GIS (Geographic Information System).
[0012]
The accumulated data includes static geographical data similar to elevation, rivers, roads, buildings, etc., and data that fluctuates dynamically depending on places similar to population distribution, occupation distribution, vegetation, rainfall, etc. There is.
[0013]
By combining static data and dynamic data, you can discover unique data patterns that can be found only in specific regions, or discover the graphical relationships that exist between multiple distribution maps. It is possible.
[0014]
For example, by comparing the three distribution maps of the vegetation distribution, rainfall, and landslide occurrence area, the annual rainfall is higher than the average near the point where landslides frequently occur, and the vegetation distribution In the figure, this is the case where a relationship such as being in an area without trees is discovered.
[0015]
In a conventional spatio-temporal database system, a relationship between figures having an area or volume as a figure can be calculated.
[0016]
However, the relationship calculation between figures having an area or a volume generally requires a very large cost. For this reason, the data handled by the spatio-temporal database system needs to be so-called static data that is updated little to the extent that it does not affect even if time is required for the calculation of the relationship.
[0017]
On the other hand, the movement trajectory data is a kind of sensor information and has a property that data is added in real time. Therefore, the movement trajectory database needs to be capable of searching data sufficiently fast even for dynamically increasing data.
[0018]
However, existing space-time database systems do not have this function.
[0019]
[3] Navigation system (for example, see Patent Document 3)
There are a car navigation system and a personal navigation system as systems capable of specifying a plurality of points on a map and searching for a route passing in a specified order.
[0020]
The navigation system has road shape, length, and connection information as data, and the car navigation system has information on traffic congestion, construction, and traffic restrictions. Find the best route that satisfies
[0021]
However, the search target of the navigation system is static road data as a route connecting two points, and movement trajectory data in which data is added in real time with an indefinite shape cannot be set as a search target.
[0022]
[Patent Document 1]
JP-A-6-96234
[Patent Document 2]
JP 7-93338 A
[Patent Document 3]
Japanese Patent Laid-Open No. 9-178499
[Non-Patent Document 1]
M.Egenhofer, E.Clementini, and P.di Felice: Topological Relati
ons between Regions with Holes, International Journal of Geog
raphical Information Systems, Vol.8, No.2, pp.129-142 (1994).
[0023]
[Problems to be solved by the invention]
As described above, the conventional technology cannot satisfy the following two requirements.
[0024]
That is, it is not possible to perform a search by giving spatial and temporal constraints on moving trajectory data, which is a line segment having a complicated shape with orientation, according to an intuitive description method.
[0025]
Further, it is impossible to search the database for movement trajectory data having a shape similar to that of the movement trajectory data given as a sample.
[0026]
The present invention has been made in view of such circumstances, and a movement trajectory is provided by applying constraint conditions in accordance with an intuitive description method, using movement trajectory data that is a line segment having a complicated shape stored in a database as a search target. It is an object of the present invention to provide a new movement trajectory data search technique that enables data search and enables search for movement trajectory data having a similar shape.
[0027]
[Means for Solving the Problems]
  In order to achieve this object, in the present invention, (1) Storage deviceUsed for retrieval of stored movement trajectory dataSearch information (link list and array)Search for movement trajectory data that is generated and stored in a memory area (memory area such as semiconductor memory or hard disk)Information studentsAnd (2) held in its memory areaSearch informationUseTo the storage deviceA movement trajectory data search device is provided for searching for data satisfying the search condition from the stored movement trajectory data.
[0028]
  [1] Moving locus data search of the present inventionInformation studentsConfiguration of the generator
  Moving locus data search of the present inventionInformation studentsDeviceIs information generated by a system including a storage device that stores movement locus data defined by time-series data of paired data of an identifier of a movement locus, coordinates of a movement locus point, and movement time. Information used to search the accumulated movement trajectory dataTo generate the process,(1)Means for generating a divided section to be a registration destination of an area to which the movement locus point belongs by dividing a section associated with the space in which the movement locus data exists;(2) Movement locus identifier and movementMovement locus pointWhenIs assigned to the divided section to which the movement trajectory point belongs.(3)Identified subdivisionAnd a resource generated in association with the identifier of the given movement trajectory.End of the listMinutes registered in the tailIf it is different from the split parcel,ThatAt the end of the link list, register the identification information of the specified divisionAnd (4) an identifier of the given movement trajectory in association with the identifier of the movement trajectory, a means for generating a link list for recording information on the area through which the movement trajectory data passes and information on the order of passage. Pointer information indicating where the pair data for a given movement locus point is stored in the storage device with respect to the array associated with the divided section registered at the end of the link list generated in association with Means to register, (5) minutesWhen the length of the array provided in association with the split partition is longer than the specified length, a split section having a hierarchical structure is generated by splitting the split area having the array, and a link list is accordingly generated. And regenerating means.
[0029]
Each of the above processing means can be realized by a computer program, and the computer program can be provided by being recorded on a recording medium such as a semiconductor memory.
[0030]
  [2] Configuration of the movement trajectory data retrieval apparatus of the present invention
  Moving locus data search device of the present inventionAre stored in a storage device that stores movement trajectory data defined by time-series data of paired data of the movement trajectory identifier, the coordinates of the movement trajectory point, and the movement time.It is a process that searches for motion trajectory data.(1) means for holding a link list generated by the movement trajectory data search information generating apparatus of the present invention; and (2) an array in which pointer information is registered by the movement trajectory data search information generating apparatus of the present invention. And (3) a time condition is described and a lineDescribe the geometric relationship between minutes and regionsThe search conditions for moving trajectory data, whose geometric relationship is the entry of the line segment into the specified area, the exit of the line segment from the specified area, the return to the specified area, the specification The test consists of one or a combination of passing through the specified area and moving between the specified areas.Means for entering search conditionsAnd (4) link listUse search criteriaThe geometric relationship between the line segment described inCheck the moving trajectory data to be satisfiedSearching and acquiring information on the movement time of the searched movement trajectory data from the storage device based on the pointer information registered in the array, and searching for them based on the acquired information on the movement time Finally, search for movement trajectory data that satisfies the time conditions described in the search conditions from the movement trajectory data.Means to(5)When the divided sections are hierarchically configured, the area specified by the search condition is the divided section.Any ofIf they do not matchNisoAnd a means for specifying the smallest divided section including the area.
[0031]
  In such a configuration, the means for searching is when the divided sections are hierarchically configured, and the area specified by the search condition is the divided section.Any ofIf they don't matchWhen using the linked list to search for movement trajectory data that satisfies the geometrical relationship between the line segment and the region described in the search condition, the link list is used.Trajectory data that satisfies the search condition in which the specified area has been replaced by the specified minimum divided sectionInspectProcess to search.
[0033]
Each of the above processing means (excluding the means for holding) can be realized by a computer program, and this computer program can be provided by being recorded on a recording medium such as a semiconductor memory.
[0034]
  [3] Moving locus data search processing realized by the present invention
  In the present invention configured as described above, first, the movement trajectory data search of the present invention is performed.Information studentsThe generation apparatus operates to divide a section associated with a space where the movement track data exists, thereby generating a divided section serving as a registration destination of the region to which the movement track point belongs.
[0035]
  continue,Trajectory identifier and transitionMovement locus pointWhenIs specified, the division block to which the movement locus point belongs is specified.AndSpecific partitionAnd a resource generated in association with the identifier of the given movement trajectory.NcrisToendRegistered inIf the division is different,ThatAt the end of the link list, register the identification information of the specified divisionAs a result, a linked list for recording the time-series data of the identification information of the divided sections through which the movement trajectory data passes is generated in association with the identifier of the movement trajectory.
  Further, for the array associated with the divided section registered at the end of the link list generated in association with the identifier of the given movement locus, the pair data regarding the given movement locus point is stored in the storage device. The pointer information indicating whether or not it has been registered is registered.
[0036]
And when the length of the array provided in association with the divided section is not less than a specified length, the divided area having the array is divided to generate a divided section having a hierarchical structure, and accordingly Regenerate the linked list.
[0037]
  Thus, the movement trajectory data search of the present inventionInformation studentsDescribes the divided sections through which the movement trajectory data passes and the order in which they pass.Link listWill be generatedIn addition, by registering pointer information to the actual data of the movement track for the array associated with the divided section registered in the link list, information on the correspondence between the link list and the actual data of the movement track is registered. Will be generated.
[0038]
  ThisLink list and correspondence informationWhen the generation request for the movement trajectory data is issued, the movement trajectory data search apparatus of the present invention operates.Describe the time conditions andDescribe the geometric relationship between minutes and regionsMoving trajectory dataIf a search condition is given, enter it.
[0039]
Subsequently, a partition that matches the area specified by the input search condition is identified, and if there is no matching partition, the smallest partition including the area specified by the input search condition is specified. To do.
[0040]
  And if you can identify the partition that matches the area specified by the input search condition, Link listSearch using it is possibleAnd link listUse search criteriaThe geometric relationship between the line segment described inCheck the moving trajectory data to be satisfiedSearching and acquiring information on the movement time of the searched movement trajectory data from the storage device based on the pointer information registered in the array, and searching for them based on the acquired information on the movement time Finally, search for movement trajectory data that satisfies the time conditions described in the search conditions from the movement trajectory data.The
[0042]
  On the other hand, if it is not possible to identify the partition that matches the area specified by the entered search conditions, Link listSince the search used cannot be performed, the area specified in the search condition is replaced with the smallest divided section including that area.In the linked listEnable useLink riskThen, the movement trajectory data satisfying the replaced search condition is searched, and the movement trajectory data satisfying the original search condition is searched from the search.
[0044]
  In this way, the present invention describes the geometric relationship between line segments and regions.InspectionTo be able to search for movement trajectory data that satisfies the search conditionsLink list forDescribes the geometric relationship between the line segment and the region because it generates and uses it to search for movement trajectory data that satisfies the search conditions.InspectionThe search condition can be specified, and the movement trajectory data which is a line segment having a complicated shape with a direction can be searched.
[0045]
As a result, according to the present invention, it is possible to search for movement trajectory data by giving a constraint condition according to an intuitive description method, and it is possible to search for movement trajectory data having a similar shape. It becomes like this.
[0046]
  In addition, what is included in the answer obtained as a result of the searchTo the storage deviceThe entire (one) moving trajectory data stored or a part thereof. SometimesIn the storage deviceThen, even if it is one piece of movement trajectory data, as an answer, a plurality of parts of the movement trajectory data may be cut out and used as an answer.
[0047]
DETAILED DESCRIPTION OF THE INVENTION
  Next, embodiments of the present invention will be described with reference to the drawings.
  Below, the link list recorded about the time-series data of the identification information of the divided sections through which the movement trajectory data has passed may be referred to as an index.
[0048]
FIG. 1 illustrates the overall configuration of a database system 1 having the present invention.
[0049]
As shown in this figure, the database system 1 of the present invention comprises an input unit 10, a data management unit 11, a search unit 12, and a storage unit 13.
[0050]
The input unit 10 determines whether the text information received from the outside is movement trajectory data or an inquiry description. If the text information is movement trajectory data, the input unit 10 shapes the data and transmits the data to the data management unit 11. If it is an inquiry description, the contents of the inquiry are analyzed and the analysis result is transmitted to the search unit 12.
[0051]
The data management unit 11 creates an index for the data received from the input unit 10. The index is held by the data management unit 11, and the movement trajectory data is transmitted to the storage unit 13.
[0052]
The storage unit 13 records the movement trajectory data received from the outside in a physical storage device such as a hard disk, reads the movement trajectory data requested from the outside from the physical storage device, transmits it to the request source, and is instructed from the outside. It has a function of deleting movement locus data or replacing data.
[0053]
In response to the query received from the input unit 10, the search unit 12 queries the data management unit 11 to select data candidates that satisfy the query using an index, and then acquires candidate data from the storage unit 13. In addition, it has a function of selecting only the data that actually satisfies the condition from them and transmitting it as a search result to the outside.
[0054]
[1] Movement trajectory data handled in the present invention
Next, movement trajectory data handled by the database system 1 of the present invention will be described.
[0055]
A movement trajectory of an object moving in space is recorded as a set of an identifier of the object, a time, and coordinates (point data) in the space is called movement trajectory data.
[0056]
The movement trajectory data is generally a discrete point data string along a time series that is not continuous. What connected the point data sequence included in the movement locus data in the order of the time of the point data becomes a line segment having a finite length in the space and a direction.
[0057]
FIG. 2 illustrates an example of movement trajectory data. As shown in this figure, each point of the movement trajectory data is represented by a set of time t, coordinates (x, y), and identifier u assigned by the system to each movement trajectory data.
[0058]
[2] Conditional expression that can be handled by the search unit 12
Next, condition description expressions that can be handled by the search unit 12 will be described.
[0059]
As a method for searching for a graphic in space, a method using a topological relationship with another graphic in space has been conventionally used.
[0060]
As a method of describing the topological relationship between closed regions (a continuous region having a boundary in space and having no holes inside), the boundary ∂R of the region R and the internal R ° of the region (the closed region) Open area that does not include the boundary) and the outer R of the area^-There is a method using a determinant representing a relationship with (a portion not included in the region R in the space) (for example, see Non-Patent Document 1).
[0061]
For example, the relationship determinant T shown in FIG.₀(R₁, R₂) Is expressed as shown in FIG. Here, “φ” means that the common part is empty (does not exist), and “¬φ” means that there is a common part.
[0062]
In the database system 1 of the present invention, this is expanded so that the relationship between one closed region R in space and the directed line segment L can be expressed as an inner region R ° / region boundary ∂R / region outer R.^-And the starting point of the line₀L / End point of line segment₁It is expressed as a 3 × 3 matrix of L / internal L ° of the line segment (a point on the line segment other than the start point and end point) so that it can be specified as a search condition.
[0063]
For example, the relationship determinant T (R) shown in FIG.₁, L) is expressed as shown in FIG. Here, “φ” means that the common part is empty (does not exist), and “¬φ” means that there is a common part.
[0064]
As shown in FIG. 5, the two-dimensional space R²Then, there are 26 possible combinations of the relationship between the closed region R and the directed line segment L. FIG. 6 illustrates the 26 types of relational determinants.
[0065]
However, these 26 types of relationships can be described by a combination of the seven types of basic relationships using the following CONNECT operation.
[0066]
The CONNECT operation * has a region R and two directed line segments L and L '.₁L = ∂₀Only when there is a relationship L ′ ”.₁, L) and T (R₁, L ′) can be defined between two relations “T (R₁, L) * T (R₁, L ′) ”is defined as shown in FIG.
[0067]
FIG. 8 shows an example of this CONNECT operation. For example, in the cases (a) and (b) shown in the figure, the same region R₁Two directed line segments L₁, L₂Each have a different relationship,₁L₁= ∂₀L₂”And“ T (R₁, L₁) * T (R₁, L₂) "Can be calculated. In the case of (d) and (e) shown in the figure, the same region R₁Two directed line segments L_Four, L_FiveEach have a different relationship,₁L_Four= ∂₀L_Five”And“ T (R₁, L_Four) * T (R₁, L_Five) "Can be calculated.
[0068]
As described above, the 26 types of relational determinants shown in FIG. 6 can be described by combinations of the seven types of basic relationships using the CONNECT operation.
[0069]
FIG. 9 illustrates the seven types of basic relationships. By using these seven types of basic relationships and the CONNECT operation, all 26 types of relationships can be expressed.
[0070]
FIG. 10 shows an example of the combination. T in the figure_I~ T_XXVICorresponds to a determinant representing each relationship in FIG.
[0071]
As described above, the relationship between the closed region and the directed line segment can be described using seven basic relationship determinants and the CONNECT operation.
[0072]
In this way, the inquiry to the database system 1 of the present invention is described in the following format.
[0073]
<L> = Q (R, E₀* E₁*_...* E_n)
Where <L> is the set of line segments L
This equation is related to region R and relationship E₀* E₁*_....* E_nIs a query for obtaining a set <L> of line segments L having
[0074]
A relational expression with a plurality of regions can be expressed by a column of a relational determinant between the line segment and each area. For example, two regions R₁, R₂Exists and R₁R starting from the boundary of₂The relationship where the end point is on the boundary of is expressed as follows.
[0075]
T₁(R₁, R₂, L) = <T₀(R₁, L₁), T₀(R₂, L₂)> = <E_D, E_E>
However, R₁° ∩R₂° = “φ”, L₁∪L₂= L, ∂₁L₁= ∂₀L₂And
[0076]
In this equation, the line segment L is first the region R.₁And E_D(And E between all other territories)_FAnd immediately after that R₂E between_ERepresents the relationship of having
[0077]
Relational inquiries for multiple areas are also written as follows, as in the case of a single area. However, each E_iIs more than one E_jMay be combined by a CONNECT operation.
[0078]
<L> = Q ((R₁, E₀), (R₂, E₁), ..., (R_n, E_n))
In addition to the conditional expression Q, the following condition description regarding time can be given. ∂₀L time is t_S, ∂₁L time is t_E, The time of any one point on the movement trajectory data is t_iIt is written.
[0079]
(I) Independent time designation P
Regardless of Q, the time interval and time in which the movement trajectory data is included are designated. “T_E-T_S<2.0 (minutes) "or" 8:00 PM <t_E<9:00 PM "or" 11:00 AM=t_iThis is the case. These are, for example,
<L> = Q∩P₀∩P₁...
As an AND condition.
[0080]
(B) Dependent time designation S
A time interval and a time are specified for each pair of Q relation matrix and region. For example,
<L> = Q ((R₁, E₀, [T_E= 6:00 PM]), (R₂, E₁))
Write like this. In this example, “Line segment is R₁And E₀When there is a relationship, the end point time of the partial movement trajectory data is 16:00 PM. Then R₂And E₁It represents the condition of “having a relationship of”.
[0081]
[3] Index generation processing in the data management unit 11
Next, an index generation process executed by the data management unit 11 will be described.
[0082]
Assume that there are constants H, W, m, n, and θ determined by the database administrator. Each point of the movement trajectory data is represented by p = (u, t, x, y) (where u is the trajectory ID, t is the time, and x and y are the positions of the points). Each point is added to the database one by one.
[0083]
When the point p is sent to the data management unit 11 to be added to the database, the data management unit 11 creates an index in the following procedure ((i) to (v)).
[0084]
However, it is assumed that the added p is always the last (the time of p is the last among the other point groups) in the point group having the same IDu.
[0085]
(I) Index vertex τ₀Is not present, as shown in FIG.₀And a rectangle having a size of H × n and W × m centered on the position associated with the position (x, y) of p is generated. Further, the rectangle is divided into vertical n and horizontal m, and each of the divided rectangles is designated as Rec (τ₀, i, j).
[0086]
This divided rectangle Rec (τ₀, i, j) are associated with the area where each point constituting the movement trajectory data is located.
[0087]
Where Rec (τ₀, 0,0) indicates the uppermost divided rectangle for convenience, and the lower right divided rectangle is designated as Rec (τ₀, (m-1), (n-1)). Τ₀Indicates that the vertex of the index is the parent node.
[0088]
(Ii) a divided rectangle Rec (τ) including the coordinates (x, y) of p₀, i, j).
[0089]
As will be described later, the divided rectangle Rec (τ₀, i, j), when the number of points exceeding the threshold θ is registered, in order to prevent the search performance from deteriorating, the divided rectangle Rec (τ₀, i, j) is divided in an n × m division form, so that it has child node sets.
[0090]
From this, if the retrieved divided rectangle has a child node set, the corresponding divided rectangle Rec (Rec (τ₀, i, j), i ′, j ′). Similarly, a child node is searched and recursively searched until a divided rectangle having no child node is reached. Let R be the last found rectangle.
[0091]
(Iii) p is added to the end of the array associated with the searched divided rectangle R.
[0092]
Specifically, since the information of p is stored in the storage unit 13, the address of the storage destination is added to the end of the array. In FIG. 11, the divided rectangle Rec (τ₀, 3, 2), an example is shown in which p is added to the end of the array.
[0093]
At this time, if the length of the array is equal to or greater than the threshold value θ, (iv) described later is executed. On the other hand, if not, a link list (recording the passing order of the divided rectangles of the movement trajectory data) is created for p.
[0094]
The procedure for creating the link list is as follows (1) to (3).
[0095]
{Circle around (1)} Find the point p ′ with the same IDu as p and the time value immediately before. If p 'does not exist, a new top node of the link list is created, and the divided rectangle R to which p belongs is added to the top of the link list, and the process ends.
[0096]
(2) If p 'exists, if R' to which p 'belongs and R to which p belongs are "R = R'", no movement is performed because the movement is within the same divided rectangle. To finish. Otherwise proceed.
[0097]
(3) Add R to the end of the link list. R is the parent node R_pIf there is "R_p= R '". "R_p= R '", the process is terminated. Otherwise, R 'parent node R'_pSimilarly for “R_p= R '_p”. This is repeated recursively until there are no parent nodes.
[0098]
As shown in FIG. 12, according to the creation procedures of (1) to (3), finally, the movement trajectory for each hierarchical level of the divided rectangles in order from the head of the IDu point sequence (movement trajectory data). A link list created by arranging the divided rectangles through which the data passed, and τ₀And a tree describing the hierarchical relationship of the divided rectangles with the top node as the top node. There is only one link list for the p group having the same ID.
[0099]
(iv) When the length of the array of the divided rectangles R is equal to or greater than the threshold value θ, the search performance is deteriorated by registering a large number of points. Therefore, as shown in FIG. Is a parent node, R is divided into vertical n and horizontal m, and these are respectively set as Rec (R, i, j) as child nodes of R. Further, all the points p included in the array of R are rearranged in each child node, and the array of R is cleared.
[0100]
(V) When the divided rectangle R is divided, a link list is generated again for each rearranged node.
[0101]
14 and 15 show the processing flow of the index creation processing described above.
[0102]
Here, the processing flow shown in FIG. 14 is a skeleton processing flow, and the processing flow shown in FIG. 15 shows detailed processing of the link list creation processing executed in step 18 of the processing flow shown in FIG.
[0103]
In the index creation process described above, the index vertex τ₀Is used, a method of creating n × m divided rectangles in advance is used, but only one rectangle of a prescribed size is created at the center, and if necessary, the vertex τ₀It is also possible to use a method of increasing the number of divided rectangles included in. With this method, memory can be saved.
[0104]
[4] Movement locus data search processing in the search unit 12
Next, the movement track data search process executed by the search unit 12 will be described.
[0105]
When there is an inquiry for movement trajectory data from the input unit 10, the search unit 12 performs a search using the index generated by the data management unit 11.
[0106]
As mentioned earlier, the search expression is
<L> = Q ((R₁, E₀), (R₂, E₁), ..., (R_n, E_n))
Is input as an inquiry for obtaining a set <L> of line segments L.
[0107]
As R to be designated at this time, first, a case where the designated R is equal to each divided rectangle created for the index will be described, and the case other than that will be described later. E in the following sentence_A, E_BAre shown in FIG._A, E_BIt corresponds to.
[0108]
Since the movement trajectory data to be searched is a discrete point sequence, there are almost no cases where the points exactly match “on the boundary of the rectangle” in the actual query. Therefore, E_ATo E_GAmong them, for a query having a start point or an end point at the boundary, it is highly possible that no answer will be obtained even if movement trajectory data that exactly satisfies this is retrieved.
[0109]
From now on, E is the basic element for practical use._ATo E_GThe five relationships shown in FIG.
[0110]
In practical terms,

It is sufficient to use these five relationships for inquiries.
[0111]
Next, a method for searching for answers to these inquiries will be described.
[0112]
(I) Enter:
When the inquiry formula is Q ((R, Eenter)), first, the movement trajectory data passing through R is searched using an index. Next, using the link list of each movement trajectory data, it is checked whether or not it passes through another divided rectangle R ′ immediately before R (unless R is not the head of the link list, the four sides of R are immediately before R). Through one of the split rectangles). If R ′ exists, the movement trajectory data is returned as an answer.
[0113]
(II) Exit:
When the inquiry formula is Q ((R, Exit)), first, the movement trajectory data passing through R is searched using an index. Next, using the link list of each movement trajectory data, it is checked whether or not it passes through another divided rectangle R ′ immediately after R (unless R is the end of the link list, the four sides of R are immediately after R). Through one of the split rectangles). If R 'exists, return its trajectory data as an answer.
(III) Return:
When the inquiry formula is Q ((R, Ereturn)), first, the movement trajectory data passing through R is searched using an index. Next, using the link list of each movement trajectory data, it is checked whether or not there is a route on the link list that follows the links in order from R and returns to R. If such a route exists, the movement trajectory data is returned as an answer.
[0114]
(IV) Cross:
When the inquiry formula is Q ((R, Ecross)), first, the condition of Eenter is checked for R, and then the case of Eexit is further checked. The trajectory data that satisfies both is returned as an answer.
[0115]
(V) Path:
The query expression is Q ((R₁, R₂, Epath)) first, R₁Search the movement trajectory data passing through using the index. Next, using the link list of each movement trajectory data, R₁Follow links in order from R₂Find out if there is a route to get to. If the route exists, the movement trajectory data is returned as an answer.
[0116]
Of the answers obtained by the above processing, the search unit 12 that examines the conditions related to the time included in the conditional expression finally returns only the movement trajectory data that also satisfies the time conditions as an answer.
[0117]
Next, a search method in the case where the region R specified by the search formula does not match each divided rectangle created for the index will be described.
[0118]
(A) R specified by the retrieval formula is τ₀If it is larger than (the index exceeds the target range), the index cannot be used, so the trajectory that satisfies the search expression by directly examining the given conditions for all the trajectory data Search for data.
[0119]
(B) When R specified by the search expression is completely included in any one of the lowermost divided rectangles of the divided rectangular hierarchy of the index, the following calculation is performed.
[0120]
(B-1) For Enter / Exit / Return / Cross:
If the lowermost divided rectangle including R is R ′, first, the trajectory data passing through R ′ is searched using the index. After that, the condition of each movement trajectory data is checked without using the link list.
[0121]
(B-2) For Path:
R₁R is the lowest division rectangle that contains₁', R₂R is the lowest division rectangle that contains₂Then, first of all, using the index, R₁'After passing through R₂Check the movement trajectory data passing through ', and then use Q ((R₁, R₂, E_path)) Is satisfied.
[0122]
(C) If R specified in the search expression is not completely included in any of the lowermost division rectangles in the index division rectangle hierarchy, a link is made by using the smallest division rectangle in the index including R. After examining the list, the movement trajectory data satisfying the search expression is searched without using the index. The checking method is the same as when R is completely included in the lowermost divided rectangle.
[0123]
A linked list is particularly useful for computing queries that include Path. For example, R₁, ..., R_nIn a Path query that includes a large number of divided rectangles that pass sequentially through the link list, the movement trajectory is R₁To R_nYou can easily check whether or not you have passed in order.
[0124]
17 to 21 show the processing flow of the movement trajectory data search process described above.
[0125]
Here, the processing flow of FIG. 17 and FIG. 18 is a processing flow, and the processing flow of FIG. 19 and FIG. 20 is the movement trajectory data executed in step 23 / step 30 of the processing flow of FIG. 17 and FIG. The detailed processing of the search processing is shown, and the processing flow of FIG. 21 shows the detailed processing of the search processing of the movement trajectory data executed in step 34 of the processing flow of FIG. 17 and FIG.
[0126]
If the processing flow of FIG.17 and FIG.18 is demonstrated easily, in step 21,22, the division | segmentation rectangle (rectangle which became the division | segmentation origin) of the highest layer of an index and the magnitude | size of the area | region R designated by a search formula will be demonstrated. If they match, according to the processing flow of FIG. 19 and FIG. 20, follow the same procedure as in (B) above, and use the index (including the link list) to satisfy the search formula. Search for data.
[0127]
If the division rectangle of the highest layer of the index (the rectangle that is the division source) does not match the size of the area R specified by the search expression, in step 25, the area R specified by the search expression If the region R specified by the search formula is larger, an index cannot be used. Therefore, as described in (A) above, no index is used. In addition, the movement trajectory data satisfying the retrieval formula is retrieved.
[0128]
On the other hand, if the region R specified by the search expression is smaller in step 25, the process is repeated from step 28 to step 33 to specify the divided rectangle and the search expression one layer below the index. It is determined whether or not the size of the region R to be matched matches, and if they match, according to the processing flow of FIG. 19 and FIG. To search for movement trajectory data that satisfies the search expression.
[0129]
Then, by repeating the processing of step 28 to step 33, when it is finally determined that the size of the region R specified by the search formula and the divided rectangle of the index do not match, the search formula is used. In step 34, the smallest divided rectangle R ′ including the region R to be processed is identified, and the divided rectangle R ′ is treated as the one specified by the search expression according to the processing flow of FIG. Then, in step 35, a search satisfying the search expression is searched from the searched movement trajectory data without using the index.
[0130]
That is, when the minimum divided rectangle R ′ including the region R specified by the search expression is specified according to the same procedure as in the above (C), the divided rectangle R ′ is treated as being specified by the search expression. A search is performed using the index, and then, the searched movement trajectory data is searched for data satisfying the search formula without using the index.
[0131]
Here, in the processing flow of FIG. 21, the search is not divided into Enter / Exit / Return / Cross even if such a condition is satisfied for the smallest divided rectangle R ′. This is because such a condition is not always satisfied for the region R.
[0132]
Note that, regardless of which search route is used, the final search result is obtained by searching for a search that satisfies the time condition specified by the search expression.
[0133]
In the embodiment described above, data on a two-dimensional plane is handled as movement trajectory data. However, in the present invention, the present invention can be applied as it is to movement trajectory data in a three-dimensional space or more. It can be done. In this case, when the number of data representing the coordinates of each point of the movement trajectory data is data in the n-dimensional space, the number is n.
[0134]
In addition, the movement trajectory data handled by the present invention is not limited to the movement trajectory data of a moving object, but also applies to movement trajectory data in which a plurality of parameters change in conjunction with the time axis. It can be done. For example, in a case where four parameters of student's height, weight, chest circumference, and sitting height are measured each year, it can be handled as movement trajectory data in a four-dimensional space with these four parameters as axes.
[0135]
【The invention's effect】
Data obtained by recording a movement trajectory or the like of an object moving in space as a set of an object identifier, time, and coordinates (point data) in the space is referred to as movement trajectory data.
[0136]
The movement trajectory data is generally a discrete point data string along a time series that is not continuous. What connected the point data sequence included in the movement locus data in the order of the time of the point data becomes a line segment having a finite length in the space and a direction. A database system having a function of accumulating and retrieving such movement locus data is called a movement locus database system.
[0137]
In the present invention, when a topological relationship between a plurality of areas in space and movement trajectory data is designated as a search condition for such a movement trajectory database system, the movement trajectory database system stores in advance. By using an index called a grid link that is created by detecting the order of grids through which the moving trajectory data passes, it is possible to quickly search for moving trajectory data that satisfies a specified search condition.
[Brief description of the drawings]
FIG. 1 is an overall configuration diagram of a database system provided with the present invention.
FIG. 2 is a diagram illustrating an example of movement trajectory data.
FIG. 3 is an explanatory diagram of a determinant describing a topological relationship between closed regions.
FIG. 4 is an explanatory diagram of a determinant describing a topological relationship between a closed region and a directed line segment.
FIG. 5 is an explanatory diagram of a relationship between a closed region and a directed line segment.
FIG. 6 is an explanatory diagram of a determinant describing a topological relationship between a closed region and a directed line segment.
FIG. 7 is an explanatory diagram of a CONNECT operation.
FIG. 8 is an explanatory diagram of a CONNECT operation.
FIG. 9 is an explanatory diagram of a basic relation matrix.
FIG. 10 is an explanatory diagram of a relation matrix derived by a basic relation matrix and a CONNECT operation.
FIG. 11 is an explanatory diagram of an index.
FIG. 12 is an explanatory diagram of a link list.
FIG. 13 is an explanatory diagram of rectangle division processing.
FIG. 14 is a processing flow executed by a data management unit.
FIG. 15 is a processing flow executed by a data management unit.
FIG. 16 is an explanatory diagram of a relationship between a closed region and a directed line segment that are fundamental in practical use.
FIG. 17 is a processing flow executed by a search unit;
FIG. 18 is a processing flow executed by a search unit.
FIG. 19 is a processing flow executed by a search unit;
FIG. 20 is a processing flow executed by a search unit.
FIG. 21 is a processing flow executed by a search unit;
[Explanation of symbols]
1 Database system
10 Input section
11 Data management department
12 Search part
13 Accumulator

Claims (14)

Information generated by a system including a storage device that stores movement track data defined by time-series data of paired data of a movement track identifier, a coordinate of a movement track point, and a movement time, and is stored in the storage device. and a movement trajectory data search information producing formation apparatus for generating information is found using the search moving locus data,
Means for generating a divided section to be a registration destination of an area to which the movement locus point belongs by dividing a section associated with the space in which the movement locus data exists;
If the identifier of the moving track and the moving locus point is given, and means for specific the divided compartments belongs its movement trajectory point,
In the case where the split pane identified above SL, a split partition to be registered in the tail end of the linked list generated in association with the identifier of the mobile trajectory given above are different, the end of the linked list, and the specific in Rukoto to register identification information of the divided compartments, in association with the identifier of the moving track information of the area through the movement trajectory data and means for generating a linked list to be recorded for its the path sequence information,
For the array associated with the divided section registered at the end of the link list generated in association with the identifier of the given movement locus, the pair data for the given movement locus point is the storage device. And means for registering pointer information indicating where it is stored ,
Movement locus data search information producing formation apparatus characterized. In the movement locus data search information producing formation apparatus according to claim 1,
When the length of the array is not less than a specified value, a means for generating a divided section having a hierarchical structure by dividing the divided area having the array and regenerating the link list accordingly. To prepare,
Movement locus data search information producing formation apparatus characterized. In the movement locus data retrieval apparatus for searching a moving locus data stored in the storage device for storing moving locus data defined by time-series data of pair data of the coordinates and the movement time of the identifier and the movement locus point of the movement locus There,
Means for holding a link list generated by the movement trajectory data search information generating device according to claim 1;
Means for holding an array in which pointer information is registered by the movement trajectory data search information generating device according to claim 1;
Together describe time condition, a search condition of the movement locus data that describes the geometric relationship between the line segments and the area, the geometrical relationship, entry segment to the area specified , exit line from the region to be specified, return to the area specified, the passage of the area specified, the search condition constituted by any one or a combination of movement between areas designated Means to input;
Using upper SL linked list, a movement locus data satisfying the geometric relationship between the line segment and the regions described above search criteria and search, based on the pointer information registered in the sequence, Information on the movement time of the searched movement trajectory data is acquired from the storage device, and is described in the search condition from the searched movement trajectory data based on the acquired information on the movement time. further comprising a final search to that means the movement locus data that satisfies the time condition,
A moving trajectory data search device as a feature. In the movement trace data search device according to claim 3,
The divided compartments in cases hierarchically configured, if the area designated by the search condition does not match any of the dividing partition, means for identifying a minimum of the divided compartment containing a region of its With
Means for the search, using the above SL linked list, in order to find the movement locus data satisfying the geometric relationship between the line segment and the regions described by the search condition, on SL minimum splitpane Searching for movement trajectory data that satisfies the above search conditions in which the specified area has been replaced,
A moving trajectory data search device as a feature. Information generated by a system including a storage device that stores movement track data defined by time-series data of paired data of a movement track identifier, a coordinate of a movement track point, and a movement time, and is stored in the storage device. generates information used for searching for movement locus data, move the locus data to be executed by the movement locus data search information generating apparatus comprises a first generating means and the identifying means and the second generating means and registration means a search for information generation forming method,
It said first generating means to divide the compartment associated with the space existing of the moving path data, generates an splitpane as a registration destination of the region belongs movement trajectory point,
The specific means, if the identifier and the moving locus point of the movement locus is given to identify the above splitpane Field of the movement locus point,
Said second generating means, a dividing partition identified above SL, when the split compartment and to be registered in the tail end of the linked list generated in association with the identifier of the mobile trajectory given above are different, the link the end of the list, in Rukoto to register identification information of the divided sections specified above, in association with the identifier of the mobile track, the link list for recording the information of the area through the movement trajectory data and its path sequence information Produces
The pair of the given movement trajectory points with respect to the array associated with the divided section registered at the end of the linked list generated by the registration means in association with the identifier of the given movement locus. Registering pointer information indicating where the data is stored in the storage device,
Movement locus data search information producing formation wherein. In the movement locus data search information producing formation method according to claim 5,
The movement trajectory data search information generating device further includes a correcting means,
Said modifying means, if the length of the upper Symbol sequence is defined above, to generate a divided compartment with a hierarchical structure by dividing the divided region having the sequence, the linked list in response thereto generated It should be noted that the scores a,
Movement locus data search information producing formation wherein. An input means and a search means for searching for movement trajectory data stored in a storage device for accumulating movement trajectory data defined by time-series data of paired data of a movement trajectory identifier, movement trajectory point coordinates and movement time a moving locus data search method performed by the movement trajectory data searching apparatus comprising,
Said input means, as well as describe the time condition, a geometrical search movement locus data that describes the relationship between the line segments and the area, the geometrical relationship, the area designated It is composed of one or a combination of the following: entry of line segment, exit of line segment from specified area, return to specified area, passage of specified area, movement between specified areas the search criteria and enter that,
It said search means, using the link list generated by the movement trajectory data retrieval information generating apparatus according to claim 1, satisfying the geometric relationship between the line segment and the regions described by the search criteria the movement locus data searches for, on the basis of the pointer information registered in the registration sequence of pointer information by moving route data retrieval information generating apparatus according to claim 1, from the storage device, and their search The movement trajectory data satisfying the time condition described in the search condition is acquired from the searched movement trajectory data based on the acquired movement time information based on the acquired movement time information. and finally search the child,
Characteristic moving trajectory data search method. The movement trajectory data search method according to claim 7 ,
The moving trajectory data search device further includes a specifying unit,
The specific means and, when the upper Symbol dividing partitions hierarchically configured, if the area designated by the search condition does not match any of the divided sections, the smallest of the containing region of its a split pane and specific,
Said search means, by using the linked list, when looking for movement locus data satisfying the geometric relationship between the line segment and the regions described above search criteria specified above SL minimum splitpane region Searching for movement trajectory data satisfying the above search condition in which is replaced,
Characteristic moving trajectory data search method. Information generated by a system including a storage device that stores movement track data defined by time-series data of paired data of a movement track identifier, a coordinate of a movement track point, and a movement time, and is stored in the storage device. and a movement trajectory data search information producing formation program used for realization of the movement locus data search information producing formation apparatus for generating information is found using the search moving locus data,
Means for generating a divided section to be a registration destination of an area to which the movement locus point belongs by dividing a section associated with the space in which the movement locus data exists;
If the identifier of the moving track and the moving locus point is given, and means for specific the divided compartments belongs its movement trajectory point,
In the case where the split pane identified above SL, a split partition to be registered in the tail end of the linked list generated in association with the identifier of the mobile trajectory given above are different, the end of the linked list, and the specific in Rukoto to register identification information of the divided compartments, in association with the identifier of the moving track information of the area through the movement trajectory data and means for generating a linked list to be recorded for its the path sequence information,
For the array associated with the divided section registered at the end of the link list generated in association with the identifier of the given movement locus, the pair data for the given movement locus point is the storage device. where as a means for registering a pointer information indicating what is accumulated, the movement trajectory data search information producing formation program for causing a computer. In the movement locus data search information producing formation program according to claim 9,
Furthermore, when the length of the array is longer than a specified length, a divided section having a hierarchical structure is generated by dividing the divided area having the array, and the link list is generated again accordingly. As a means, including a program for operating a computer,
Movement locus data search information generation formation program which is characterized. Movement locus data retrieval apparatus for retrieving a moving locus data stored in the storage device for storing moving locus data defined by time-series data of pair data of the coordinates and the movement time of the identifier and the movement locus point of the movement locus A movement trajectory data search program used for realization,
Together describe time condition, a search condition of the movement locus data that describes the geometric relationship between the line segments and the area, the geometrical relationship, entry segment to the area specified , exit line from the region to be specified, return to the area specified, the passage of the area specified, the search condition constituted by any one or a combination of movement between areas designated Means to input;
Using the link list generated by the movement trajectory data retrieval information generating apparatus according to claim 1, the movement locus data satisfying the geometric relationship between the line segment and the regions described by the search criteria searches, based on the pointer information registered in the registration sequence of pointer information by moving route data retrieval information generating apparatus according to claim 1, from the storage device, the movement of the movement locus data those search Acquire time information, and finally search for movement trajectory data satisfying the time condition described in the search conditions from the searched movement trajectory data based on the acquired movement time information. A moving locus data search program for causing a computer to function as a means for operating. In the movement track data search program according to claim 11 ,
Furthermore, the dividing partition in cases hierarchically configured, if the area designated by the search condition does not match any of the divided compartments, identifying a minimum of the divided compartment containing a region of its Including a program for causing a computer to function,
And the search means uses the link list, in order to find the movement locus data satisfying the geometric relationship between the described are line segments and the area in the search condition, the upper SL minimum splitpane To search for movement trajectory data that satisfies the above search conditions in which the specified area has been replaced with
Characteristic trajectory data search program. A recording medium recording a moving locus data search information producing formation program according to claim 9 or 10. The recording medium which recorded the movement locus | trajectory data search program of Claim 11 or 12 .

Images