EP4544450A1 - Verfahren zum trainieren eines neuronalen netzes zum klassifizieren von dateien in dateiklassen, verfahren zum klassifizieren von dateien in dateiklassen sowie dateiklassifizierungseinrichtung - Google Patents
Verfahren zum trainieren eines neuronalen netzes zum klassifizieren von dateien in dateiklassen, verfahren zum klassifizieren von dateien in dateiklassen sowie dateiklassifizierungseinrichtungInfo
- Publication number
- EP4544450A1 EP4544450A1 EP23753806.1A EP23753806A EP4544450A1 EP 4544450 A1 EP4544450 A1 EP 4544450A1 EP 23753806 A EP23753806 A EP 23753806A EP 4544450 A1 EP4544450 A1 EP 4544450A1
- Authority
- EP
- European Patent Office
- Prior art keywords
- file
- neural network
- bif
- files
- class
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/50—Monitoring users, programs or devices to maintain the integrity of platforms, e.g. of processors, firmware or operating systems
- G06F21/55—Detecting local intrusion or implementing counter-measures
- G06F21/552—Detecting local intrusion or implementing counter-measures involving long-term monitoring or reporting
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/09—Supervised learning
Definitions
- the invention relates to a method for training a neural network to classify files into file classes, a method for classifying files into file classes and a file classification device.
- Information and/or functional security regularly faces the task of classifying files.
- malicious code can be hidden in a file using a protection tool.
- a detected previous application of a protection tool to a file can therefore indicate a compromise in IT security.
- a classification of files can reveal in particular whether such a protection tool was used or not.
- a file can be protected with security tools or left unprotected.
- such classification into file classes is regularly necessary.
- Classifying files also referred to as classifying files in the context of the present application, means the assignment of a file to exactly one of two or more predetermined file classes. In the case of exactly two file classes, this is called binary classification, otherwise it is called multinomial classification.
- file classes can either form endpoints of a continuous scale, such as a continuous scale between the "large file” and "small file” endpoints in file length classification, with in this example a clear, unambiguous classification only in the end ranges of the continuous scale is possible.
- a clear classification can also be made based on information about the presence or absence of a certain property: a property is either present or the property is not present.
- Such a property can be the property “File contains ASCII text”, which is either present or not present.
- the file classes can be classified into more than two file classes. Preselected file types can form such file classes, such as the file types “Excel file”, “PDF file” and “JPEG file”.
- neural networks can also be used to classify files. But neural networks also rely on the recognition of such strings with continued training, so that the use of neural networks does not have any significant advantage in this respect.
- test files are used, which are each assigned to a file class, with the test files being broken down into bit sequences that remain assigned to the previously assigned file class and the neural network being trained with the bit sequences.
- neural networks can be advantageously used, unlike what was previously known, for a very flexible recognition of file classes by training neural networks appropriately with test files.
- the test files i.e. H. the training files
- the neural network is specifically trained to classify the contents of the test files based on general structural features, such as “a comma occurs in the file at more or less regular intervals”.
- the neural network can therefore Do not rely on rigid patterns, which can no longer or at least no longer be reliably recognized when this pattern changes, for example as a result of an adaptation of a file format as a result of new file versions or new file standards.
- the classification rather remains stable against changes that at a point in time after training in the file formats, as long as at least the basic structure of the file contents is preserved.
- a conventional classification on the other hand, which only looks for the string '%PDF' at the beginning of the file, only works perfectly for as long as the string appears there in exactly this form. However, if this string is only changed minimally in a new version of the format or moved to a different location, it will work Detection rule no longer exists at all.
- a first neural network can therefore be trained in a significantly improved manner compared to the prior art.
- the neural network trained in this way is advantageously designed for flexible classification that is less susceptible to adjustments to file formats.
- the neural network can therefore be trained significantly more efficiently using the method according to the invention and the neural network trained using the method according to the invention functions significantly more error-free.
- the training files are completely broken down into bit sequences.
- each bit of the training files is expediently included in at least one bit sequence with which the neural network is trained. In this way, all information contained in the training files is used to train the neural network.
- the neural network is expediently trained with the bit sequences in a disordered sequence, in particular in a randomized sequence. This prevents the neural network from interpreting the succession of the bit sequences as information and consequently from assigning this sequence of bit sequences a meaning that can influence the classification by means of the neural network.
- the disordered, in particular randomized, sequence the sequence does not form a pattern that can be evaluated by the neural network. Instead, the neural network is trained to use more general structural features that can be recognized in the bit sequence for classification.
- the neural network is a recurrent first neural network, ie such a first neural network, which is known in English as a “recurrent neural network”.
- a “recurrent neural network” has a type of “memory”. and thus has the possibility of a previous one Number of bits of a bit sequence to be included in the classification decision for the current bit sequence. This makes it possible in particular to apply the neural network to bit sequences of any length, instead of being limited from the outset to bit sequences of fixed length due to its input format.
- bit sequences during training would require the network to independently learn any structural connection between successive bits for each position within the fixed block length, whereas when using recurrent networks there is a priori connection with the local context, i.e. with the “history.” ” of the last bit sequences is learned, regardless of the exact position of the current bit sequence within the file.
- a first neural network is used, which is or is being trained according to a method according to the invention for training a neural network for classifying files into file classes, as described above, and it will be The method uses at least two different excerpts from the file in the form of bit sequences and the bit sequences are each classified into candidate file classes using the first neural network and a file class is determined depending on the candidate file classes. Since, according to the invention, a first neural network trained by means of the method according to the invention for training a first neural network for classifying files into file classes is used, the method according to the invention advantageously enables a flexible, i.e. H. Classification of files that is robust to adjustments to a respective file format is possible.
- the files are preferably completely broken down into bit sequences. This means that all information contained in a file is used to classify the file. The reliability of the class sification of the file is further improved in this development of the method according to the invention.
- the classification is carried out for several different bit sequences of at least one file, the bit sequences taken together preferably forming the at least one file.
- the information content of the file can be better utilized.
- the entire available content of the file is used for classification.
- the candidate file classes into which the bit sequences are classified are determined together with the position of the respective bit sequence within the file. In this way, the position of the bit sequence can be used as additional information when finally classifying the file.
- the file class i.e. H. the end result of the classification method based on the candidate file classes is suitably determined in such a way that, for each file, the candidate file class into which the most bit sequences are classified is determined as the file class for this file.
- the file class is preferably determined in such a way that, for each candidate file class, a measure of whether the bit sequence belongs to this candidate file class, which the first neural network assigns to the bit sequence, is determined and is used to determine the file class.
- Such a measure of whether the bit sequence belongs to this candidate file class can also be referred to as a classification value.
- the classification value usually forms a number between 0 and 1 and indicates a clear assignment to the respective candidate file class, for example using a value closer to 0 or closer to 1, while an unclear assignment is often expressed using values close to 0.5.
- Such a measure for the membership of the bit sequence to this candidate file class can preferably be determined and used in the form of an average value, in particular in the form of an arithmetic mean or a geometric mean or a square mean, for the several bit sequences of the file.
- a course of this measure for the membership of bit sequences to the candidate file class can be determined along their position within the file and used to determine the file class. In particular, an accumulation of values of the measure at certain positions of the bit sequences and/or local deviations and/or statistical abnormalities of values of this measure can be determined based on the course and used to determine the file class.
- the file class is determined by means of a second neural network, to which, for each bit sequence, preferably the candidate file class into which it has been classified by the first neural network and / or for each candidate file class a measure of the membership of the bit sequence
- This candidate file class which assigns the first neural network to the bit sequence, preferably together with the position, is transmitted as input data.
- the file classification device has at least one first neural network, which has been trained by means of a method according to the invention for training a neural network for classifying files into file classes, as described above, and/or which is designed to have at least one method according to the invention for classifying a file in file classes as described above.
- the file classification device is used to use an IT security property of files, which has an impact on data and / or functional security of an automation system using the files, as a file class and depending on the File class into which the file classification device classifies the file to take an IT security measure.
- the automation system is particularly preferably a manufacturing system.
- the IT security measure is expediently providing the file with a desired IT security property if the classification using the file classification device shows that the file does not have the desired IT security property or a restriction of access rights when using or executing the file or a strengthened one Monitoring the use or execution of the file if it does not have the desired IT security properties.
- malicious code may be obfuscated in a file using a protection tool.
- a detected previous application of a protection tool to a file can therefore mean a compromise in IT security. Consequently, in a particularly advantageous development of the invention, it is viewed as the desired IT security property of a file that such a protection tool has not been previously applied to this file.
- the invention can alternatively also be described as a method for increasing the IT security of an automation and/or manufacturing system, in which a security property of the file is determined by means of the classification of the file and based on the security property determined in the automation and/or An IT security measure is taken at the manufacturing plant.
- the classification takes place tion of the one or more files by means of a method according to the invention for classifying at least one file into file classes as described above and/or by means of a file classification device according to the invention as described above, the method having an IT security feature the one or more files is used as one of the file classes and, depending on the file class into which the one or more files are classified, an IT security measure is taken.
- the IT security is expediently a functional security and/or a data security and/or the IT security property has an impact on the data and/or functional security of the automation system using the one or more files.
- the one or more files are configuration files and/or control files of the automation and/or manufacturing system.
- the security measure is to provide the one or more files with a desired IT security feature as described above or to restrict access rights when using or executing the one or more files or to increase monitoring the use or execution of one or more files if they do not have the desired IT security properties.
- malicious code may be hidden in a file using a protection tool. A detected previous application of a protection tool to a file can therefore mean a compromise in IT security.
- Fig. 1 in a flow diagram for the process of the method according to the invention for training a neural network for classifying file types in a schematic drawing and
- Fig. 2 shows a file classification device according to the invention schematically in a schematic diagram.
- a first neural network NN is trained to classify file types.
- training data TRAINO, TRAIN1 are passed to the neural network NN as input data, for which the neural network NN is to determine and output an associated classification in the form of file classes CLAO, CLA1 as the respective output data.
- the neural network NN should be able to sort unknown files FIL (see Fig. 2) into the file classes CLAO, CLA1.
- the neural network is created with the help of a training data set TRAINO from training files x ul , x u2 , x u3 , which are all assigned to the file class CLAO, and with the help of a training data set TRAIN1 from training data x pl , x p2 , assigned to the file class CLA1.
- Each training example therefore consists of a training file x ul , x u2 , x u3 , x pl , x p2 , x p3 and a label in the form of the file class CLAO, CLA1, where the file class CLAO, CLAO has a value of 0 or 1 in the illustrated exemplary embodiment and specifies the file class to which a training file x ul , x u2 , x u3 , x pl , x p2 , x p3 must be assigned if the classification is correct.
- the procedure for training the neural network NN such as withholding a small part of the training data for validation purposes, follows usual standards in a manner known per se and will not be explained further here.
- a previous application of a protection tool to an executable binary file is to be recognized.
- Using the protection tool would allow malicious code to be hidden in the file, potentially posing a security risk to a device or system on which the file is intended to be used.
- the training data sets TRAINO, TRAIN1 are provided in such a way that the protection tool is not initially applied to a data set x lr Binaries result and on the other hand the protection tool is applied to the data set x 2 , x 2 , x 3 , which creates the protected training files x pl , x p2 , x p3 , which also form binaries.
- the training data set TRAINO therefore includes the training files x ul , x u2 , x u3 together with the file class CLAO with the label "0", indicating that these are unprotected training files.
- the training data set TRAIN1 includes the training data x pl , x p2 , x p3 together with the file class CLA1 with the label “1” (protected).
- the training data sets TRAINO, TRAIN1 can also be obtained in a different way.
- the training data set of a file class CLA1 it is not necessary for the training data set of a file class CLA1 to be based on the training data set of the other file class CLAO.
- the special feature of the method presented lies in the preparation of the training files x ul , x u2 , x u3 , x pl , x p2 , x p3 in such a way that the neural network NN strengthens its judgment on structural properties of the file contents of the training files x ul , x u2 , x u3 , x pl , x p2 , x p3 has to align and cannot rely on very specific patterns of the certain file types CLAO, CLA1, such as "the third byte always has the value 17".
- All training files x ul , x u2 , x u3 , x pl , x p2 , x p3 from the common training data set T are each divided into shorter bit sequences BIF.
- the division can be done with a uniform length, here for example 1 kilobyte, or with a variable length for each bit sequence BIF. If the division at the end of a training file x u2 , x u2 , x u3 , x p2 , x p2 , x p3, for example, can be discarded or filled with dummy bits, for example zero bits.
- the shortened bit sequences BIF together with the labels for the file classes CLAO, CLA1 belonging to their respective training files x ul , x u2 , x u3 , x pl , x p2 , x p3 , from which the bit sequences BIF come from, are added to the set of final training pairs assembled;
- the set of final training pairs therefore contains all pairs of bit sequences BIF and labels for the file classes CLAO, CLA1 for all training examples of the common training data set T.
- the order in which the bit sequences BIF and the associated labels are presented to the neural network during training becomes then - as usual - randomized at the beginning of each training epoch.
- the main advantage is that the neural network NN usually does not present a single file "as a whole". rather, the network receives individual files only limited to the length of the individual bit sequences BIF and not in the correct order, but rather in any order. Therefore, the neural network NN cannot rely on the presence of specific values at certain positions of a file to classify the files and must instead use more general structural features that can be seen in the section of a bit sequence BIF for the classification.
- the neural network NN is presented with a larger number of labels, i.e. assigned file classes CLAO, CLA1, with the same total amount of input data, namely one label per bit sequence BIF instead of one per file Data set x lr x 2 , x 3 .
- the bytes within the bit sequences BIF are not passed to the input layer of the neural network NN both during training and in later use as ordinal values between 0 and 255, but as a sequence of 8 bits each with values between 0 and 1. Otherwise, distortions would arise because higher-order bits of a byte receive an excessive share of the activation of an input neuron and, conversely, lower-order bits receive too little. Patterns in low-value bits that are significant in and of themselves could then be lost in the high random noise of higher-value bits.
- the neural network NN is a so-called “recurrent neural network”, also referred to as a recurrent network, which has a type of “memory” and thus the possibility of including previous bytes BIF of a bit sequence in the classification decision for the current byte.
- recurrent neural network also referred to as a recurrent network
- bit- Following BIF of fixed length during training means that the neural network has to independently learn every structural connection between successive bits of the file for each position within the specified block length, while when using a recurrent neural network NN there is a connection with the local context from the outset, ie the “history” of the last bytes is learned, regardless of the exact position of the current byte.
- the exemplary embodiment shown can be expanded in a further exemplary embodiment in such a way that the application of the neural network NN to the input data in the form of the file of the data record EIL provides a continuous classification process OB(t) over the entire file FIL, in which the output of the neural Network is held at position t after each byte BIF of the file FIL.
- the input data of the neural network NN can consist of a different number of bits, for example consisting of only a single bit instead of 8 bits for a byte BIF, so that even a bit-precise output OB (t). where t denotes the bit position within the file FIL.
- a final classification function f (OB (t) ) is then applied to the classification curve OB(t) determined by the network for the measure of the membership of the bit sequence to this candidate file class, which is determined for each bit sequence BIF within the file FIL.
- This classification function f (OB (t) ) can be realized, for example, as an averaging in the manner of an arithmetic mean or in the identification of a specific local cluster of strongly deviating values of the classification process or in the identification of a local cluster of similar values of the classification process.
- the classification can also be implemented using a further neural network NN2 (see FIG. 2): For example, the further neural network NN2 is trained to to assign the correct label CLAO, CLA1 to the classification history OB (t) output by the first neural network NN.
- the classification device CLAS shown in FIG. 2 uses the training of the first neural network NN described in the previous exemplary embodiments, which transfers the classification process OB(t) to the second neural network NN2.
- the second neural network NN2 determines the final file class from this, CLAO in the case shown, and transmits the file class CLAO to an action device MEA.
- the files FIL which are supplied to the first neural network NN for classification, are read from a memory of a manufacturing device MANU of a manufacturing plant.
- the MANU manufacturing facility uses the FIL files for production control.
- the MEA action facility takes action depending on the file class determined, i.e. H. depending on the judgment of the classification facility CLAS, an application of a protection function to the file FIL. If the classification device CLAS determines that a protection tool has been applied to the FIL file, which could potentially conceal malicious code introduced into the FIL file, the protection function is applied.
- the protective function is provided by isolating the FIL file in a secure environment and intensive monitoring of the use of the FIL file. If it is detected that no protection tool has been applied to the FIL file, the security function will not be used. Accordingly, the invention can alternatively be described as an application or method for increasing the IT security of a manufacturing plant.
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Software Systems (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Computational Linguistics (AREA)
- Artificial Intelligence (AREA)
- Mathematical Physics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Computer Security & Cryptography (AREA)
- Computer Hardware Design (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
Description
Claims
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| EP22187801.0A EP4312154A1 (de) | 2022-07-29 | 2022-07-29 | Verfahren zum trainieren eines neuronalen netzes zum klassifizieren von dateien in dateiklassen, verfahren zum klassifizieren von dateien in dateiklassen sowie dateiklassifizierungseinrichtung |
| PCT/EP2023/070613 WO2024023104A1 (de) | 2022-07-29 | 2023-07-25 | Verfahren zum trainieren eines neuronalen netzes zum klassifizieren von dateien in dateiklassen, verfahren zum klassifizieren von dateien in dateiklassen sowie dateiklassifizierungseinrichtung |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| EP4544450A1 true EP4544450A1 (de) | 2025-04-30 |
Family
ID=82786712
Family Applications (2)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| EP22187801.0A Withdrawn EP4312154A1 (de) | 2022-07-29 | 2022-07-29 | Verfahren zum trainieren eines neuronalen netzes zum klassifizieren von dateien in dateiklassen, verfahren zum klassifizieren von dateien in dateiklassen sowie dateiklassifizierungseinrichtung |
| EP23753806.1A Pending EP4544450A1 (de) | 2022-07-29 | 2023-07-25 | Verfahren zum trainieren eines neuronalen netzes zum klassifizieren von dateien in dateiklassen, verfahren zum klassifizieren von dateien in dateiklassen sowie dateiklassifizierungseinrichtung |
Family Applications Before (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| EP22187801.0A Withdrawn EP4312154A1 (de) | 2022-07-29 | 2022-07-29 | Verfahren zum trainieren eines neuronalen netzes zum klassifizieren von dateien in dateiklassen, verfahren zum klassifizieren von dateien in dateiklassen sowie dateiklassifizierungseinrichtung |
Country Status (4)
| Country | Link |
|---|---|
| US (1) | US20260037773A1 (de) |
| EP (2) | EP4312154A1 (de) |
| CN (1) | CN119698600A (de) |
| WO (1) | WO2024023104A1 (de) |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9721097B1 (en) * | 2016-07-21 | 2017-08-01 | Cylance Inc. | Neural attention mechanisms for malware analysis |
-
2022
- 2022-07-29 EP EP22187801.0A patent/EP4312154A1/de not_active Withdrawn
-
2023
- 2023-07-25 EP EP23753806.1A patent/EP4544450A1/de active Pending
- 2023-07-25 US US19/099,547 patent/US20260037773A1/en active Pending
- 2023-07-25 WO PCT/EP2023/070613 patent/WO2024023104A1/de not_active Ceased
- 2023-07-25 CN CN202380057373.XA patent/CN119698600A/zh active Pending
Also Published As
| Publication number | Publication date |
|---|---|
| EP4312154A1 (de) | 2024-01-31 |
| US20260037773A1 (en) | 2026-02-05 |
| WO2024023104A1 (de) | 2024-02-01 |
| CN119698600A (zh) | 2025-03-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| DE2909153C2 (de) | Einrichtung zur digitalen Analyse von Bild- oder Zeichenmustern | |
| EP3267625B1 (de) | Verfahren zur detektion von anomalen zuständen in einem computernetzwerk | |
| DE2712575C2 (de) | Assoziatives Speichersystem in hochintegrierter Halbleitertechnik | |
| DE102014204830A1 (de) | Computerimplementierte Systeme und Verfahren zum Vergleichen und Assoziieren von Objekten | |
| DE102014204834A1 (de) | Computerimplementierte Systeme und Verfahren zum Vergleichen und Assoziieren von Objekten | |
| DE10134229A1 (de) | Verfahren und System zum Ermitteln von Abweichungen in Datentabellen | |
| DE102014208212A1 (de) | Ableiten eines gerätespezifischen Wertes | |
| DE69126720T2 (de) | Schnelle Bestimmung von Unterzusammenhängen in einer Erbhierarchie | |
| DE102014207874A1 (de) | Verfahren zur automatisierten Erstellung eines eine technische Zeichnung charakterisierenden Datensatzes | |
| EP4544450A1 (de) | Verfahren zum trainieren eines neuronalen netzes zum klassifizieren von dateien in dateiklassen, verfahren zum klassifizieren von dateien in dateiklassen sowie dateiklassifizierungseinrichtung | |
| AT522281B1 (de) | Verfahren zur Charakterisierung des Betriebszustands eines Computersystems | |
| AT521665B1 (de) | Grammatikerkennung | |
| DE102019004944A1 (de) | Steuersystem einer industriemaschine | |
| DE202024104811U1 (de) | Ein System zur Klassifizierung von Reiskörnern auf der Grundlage eines neuronalen Netzes zur Auswahl von Merkmalen | |
| DE3113189C2 (de) | Vorrichtung zur Umsetzung von digitalen Zeichencodes, die von einem Datenverarbeitungssystem empfangen oder geliefert werden | |
| DE19645057C2 (de) | Vorrichtung zur Selektion von Adressenwörtern mittels Demultiplex-Decodierung | |
| DE102005008844B4 (de) | Verfahren zur computergestützten Klassifizierung von Daten und Vorrichtung zu dessen Durchführung | |
| EP4657292B1 (de) | Verfahren zur konfiguration eines erkennungssystems zur charakterisierung anomaler betriebszustände eines computersystems | |
| DE102009016588A1 (de) | Verfahren zur Ermittlung von Textinformationen | |
| DE19635351C2 (de) | Verfahren zur Formatkonvertierung | |
| DE202024002630U1 (de) | System zum Vergleichen von zwei Zeichenfolgen und computer-lesbares Speichermedium | |
| WO2026077503A1 (de) | Computerimplementiertes verfahren zum vergleichen von zwei zeichenfolgen computer-lesbares speichermedium und system | |
| DE202023101247U1 (de) | Ein System zur Vorhersage von Unfällen und Warnung der Fahrer mit einer Maschinelerntechnik | |
| AT387099B (de) | Einrichtung zur digitalen analyse von bildoder zeichenmustern | |
| DE102021131946A1 (de) | Erfassungseinrichtung, Verfahren sowie System zur visuellen Erfassung von Objekten |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: UNKNOWN |
|
| STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: THE INTERNATIONAL PUBLICATION HAS BEEN MADE |
|
| PUAI | Public reference made under article 153(3) epc to a published international application that has entered the european phase |
Free format text: ORIGINAL CODE: 0009012 |
|
| STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: REQUEST FOR EXAMINATION WAS MADE |
|
| 17P | Request for examination filed |
Effective date: 20250122 |
|
| AK | Designated contracting states |
Kind code of ref document: A1 Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC ME MK MT NL NO PL PT RO RS SE SI SK SM TR |
|
| DAV | Request for validation of the european patent (deleted) | ||
| DAX | Request for extension of the european patent (deleted) | ||
| STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: EXAMINATION IS IN PROGRESS |
|
| 17Q | First examination report despatched |
Effective date: 20251210 |