A method, system and storage medium for recognizing characters on a digital tube
By developing a character recognition method for digital tube displays based on brightness distribution and Retinex theory, the problem of character recognition accuracy under varying lighting conditions and resource-constrained environments has been solved. This method achieves efficient and robust character recognition, making it suitable for embedded devices and smart manufacturing scenarios.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- KGE
- Filing Date
- 2025-12-24
- Publication Date
- 2026-05-26
AI Technical Summary
Existing technologies suffer from low accuracy and poor robustness when recognizing characters on digital tube LED displays. They are particularly difficult to use effectively in complex scenarios such as changes in lighting and character defects. Furthermore, it is difficult to deploy deep learning models when equipment resources are limited.
By employing feature extraction based on brightness distribution and the Retinex theoretical illumination compensation algorithm, combined with image preprocessing and feature matching, we can achieve efficient and robust recognition of characters on digital tubes. Through vertical projection analysis and local adaptive threshold segmentation, we can reduce the impact of illumination changes and avoid binarization errors, making it suitable for low computing power and low storage environments.
It achieves efficient and accurate recognition of digital tube characters under complex lighting conditions, reduces computing resource requirements, improves recognition rate, is suitable for embedded devices, eliminates errors in manual data acquisition, and supports intelligent manufacturing and data analysis.
Smart Images
Figure CN122090466A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of LED digital tube recognition technology, and in particular to a digital tube character recognition method, system and storage medium. Background Technology
[0002] Digital tube LED displays are widely used in industrial instruments, home appliances, medical equipment, energy systems, and other applications due to their low cost, high brightness, and long lifespan. Currently, the acquisition of characters for digital tube LED displays mainly relies on manual acquisition, adding acquisition modules to interface with various manufacturers' protocols and use programs via communication ports, or using other image recognition devices on the market. However, this approach has the following problems: Traditional OCR technology is primarily designed for printed characters, but it is poorly adaptable to the unique discrete stroke structure, brightness characteristics, and interference factors of digital tubes, resulting in low recognition accuracy. Template matching-based methods are sensitive to changes in the size, angle, font, and brightness of digital tubes, exhibiting poor robustness and requiring the pre-production of numerous templates. Deep learning-based methods require extensive labeled data for training, resulting in complex models, high computational resource consumption, and difficulties in deployment on embedded devices or in scenarios with high real-time requirements. Their generalization ability may be insufficient for specific digital tube style variations (such as different manufacturers or different aging levels). Adding an external acquisition module requires developing programs to adapt to different manufacturers' communication protocols, increasing development workload and failing to achieve universality. Furthermore, some older devices lack external communication ports, making data acquisition impossible through an external acquisition module.
[0003] Therefore, to overcome the technical bottlenecks in data acquisition for digital tube and LED display characters, it is urgent to develop a recognition algorithm suitable for embedded devices and other resource-constrained environments. This algorithm must be lightweight, have high real-time performance, and strong robustness, enabling stable operation under low-computing-power and low-storage hardware conditions. It should effectively cope with complex scenarios such as changes in lighting, character defects, and display jitter, thereby achieving efficient and accurate data acquisition and recognition. Summary of the Invention
[0004] Based on this, it is necessary to address the aforementioned technical problems of low recognition rate of digital tubes, the need for a large amount of labeled data for deep learning training, poor binarization effect under ambient light interference leading to character breakage or adhesion, and uneven brightness affecting the accuracy of feature extraction. A digital tube character recognition method, system, and storage medium should be provided that can improve the recognition rate of displayed characters, provide robustness under ambient light interference, and ensure stable operation under low computing power and low storage hardware conditions.
[0005] To solve the above-mentioned technical problems, the technical solution of the present invention is as follows: In a first aspect, the present invention provides a method for recognizing characters on a digital tube, comprising: Obtain the original image containing the digital tube display area; The original image is preprocessed to select digital tube display areas that meet preset conditions; The digital tube display area is segmented into single characters to obtain at least one single character image; Stroke feature vectors of the single character image are extracted based on brightness distribution; The stroke feature vector is identified based on a preset matching rule to obtain at least one single character result corresponding to a single character image. All the identified single character results are combined to obtain the character recognition results of the digital tube display area.
[0006] Secondly, the present invention also provides a digital tube character recognition system, utilizing a digital tube character recognition method as described above, comprising: The image acquisition module is used to acquire the original image containing the digital tube display area; The image preprocessing module is used to preprocess the original image and filter out the digital tube display areas that meet the preset conditions. A character segmentation module is used to segment the digital tube display area into single characters to obtain at least one single character image; The feature extraction module is used to extract the stroke feature vector of the single character image based on the brightness distribution; The character recognition module is used to recognize the stroke feature vector based on a preset matching rule to obtain a single character result corresponding to at least one single character image; The character combination module is used to combine all the recognized single character results to obtain the character recognition results of the digital tube display area.
[0007] Thirdly, the present invention also provides a computer-readable storage medium storing at least one instruction, at least one program, code set, or instruction set, wherein the at least one instruction, at least one program, code set, or instruction set is loaded and executed by a processor to implement a digital tube character recognition method as described above.
[0008] Compared with the prior art, the beneficial effects of the technical solution of the present invention are: This invention utilizes the display mechanism of digital tubes to perform feature analysis and processing on the digital tube area in the acquired image, thereby achieving accurate character recognition and significantly improving the recognition rate. By locating the digital tube area, vertical projection analysis is performed to analyze the trough positions of the projection histogram, using these troughs as dividing points between characters, eliminating the need for extensive training with labeled data. Directly utilizing brightness statistics avoids errors introduced by the binarization process, significantly improving robustness to changes in illumination, slight blurring, and uneven brightness. A Retinex-based theoretical illumination compensation algorithm is used to initially mitigate the impact of non-uniform illumination. It does not rely on high-performance computing units or large-capacity storage, allowing it to be embedded in small industrial control computers or core boards. It eliminates manual copying and input, avoiding secondary errors, and achieves full automation from data acquisition to analysis, providing a solid data foundation for intelligent manufacturing, quality traceability, and data analysis. Attached Figure Description
[0009] Figure 1 This is a flowchart illustrating a digital tube character recognition method in some embodiments of this application; Figure 2 The diagram shows the process details of a digital tube character recognition method in some embodiments of this application. Figure 3 This is a flowchart illustrating a digital tube character recognition system in some embodiments of this application. Detailed Implementation
[0010] The terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such terms are interchangeable where appropriate; this is merely a way of distinguishing objects with the same attributes in the description of embodiments of this application. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion, so that a process, method, system, product, or apparatus that comprises a series of units is not necessarily limited to those units, but may include other units not explicitly listed or inherent to those processes, methods, products, or apparatuses. The term "determine" broadly covers a wide variety of actions, including acquiring, calculating, processing, deriving, investigating, searching (e.g., searching in a table, database, or other data structure), probing, and similar actions; it may also include receiving (e.g., receiving information), accessing (e.g., accessing data in memory), and similar actions; it may also include generating, creating, establishing, and similar actions; and parsing, selecting, choosing, and similar actions, etc. Definitions of other terms will be given in the following description.
[0011] It should be noted that when one element is considered to be "connected" to another element, it can be directly connected to the other element or connected to the other element through an intermediary element. Furthermore, in the following embodiments, "connection" should be understood as "electrical connection," "communication connection," etc., if there is transmission of electrical signals or data between the connected objects.
[0012] It should be emphasized that the acquisition, transmission, storage, use, and processing of data in the technical solutions of this application all comply with the relevant provisions of national laws and regulations.
[0013] In the embodiments of this application, certain software, components, models and other existing solutions in the industry may be mentioned. These should be regarded as exemplary and are only intended to illustrate the feasibility of implementing the technical solution of this application. However, they do not mean that the applicant has used or necessarily used the solution.
[0014] The accompanying drawings are for illustrative purposes only and should not be construed as limiting the scope of this patent. To better illustrate this embodiment, some parts in the accompanying drawings may be omitted, enlarged, or reduced, and do not represent the actual product dimensions; It will be understood by those skilled in the art that certain well-known structures and their descriptions may be omitted in the accompanying drawings.
[0015] The technical solution of the present invention will be further described below with reference to the accompanying drawings and embodiments.
[0016] Example 1 This embodiment provides a method for recognizing characters on a digital tube. (See attached document.) Figure 1 ,include: Obtain the original image containing the digital tube display area; The original image is preprocessed to select digital tube display areas that meet preset conditions; The digital tube display area is segmented into single characters to obtain at least one single character image; Stroke feature vectors of the single character image are extracted based on brightness distribution; The stroke feature vector is identified based on a preset matching rule to obtain at least one single character result corresponding to a single character image. All the identified single character results are combined to obtain the character recognition results of the digital tube display area.
[0017] The core of this method lies in fully utilizing the self-illuminating and discrete stroke characteristics of LED digital tubes. Through a series of targeted image processing and feature extraction steps, it achieves efficient and robust character recognition. It is suitable for scenarios such as smart meter reading, industrial dashboard recognition, and parking lot timer reading.
[0018] In this embodiment, a camera connected to the Raspberry Pi is used to acquire the original image containing the digital tube display area.
[0019] In some preferred embodiments, the original image is preprocessed, see [reference]. Figure 2 Specifically, it includes: The original image is adaptively illuminated using a Retinex-based theoretical illumination compensation method to obtain a grayscale image. Edge detection, morphological operations, and contour finding are used to locate the ROI (Region of Interest) of the grayscale image and filter out candidate digital tube display areas that meet preset conditions; The candidate digital tube display area is corrected to obtain a digital tube display area that meets the preset conditions; The preset conditions include any one or more of aspect ratio, area range, and density.
[0020] In this embodiment, a theoretical illumination compensation algorithm based on Retinex is used to initially reduce the influence of non-uniform illumination, providing a higher quality grayscale image for subsequent steps and solving the problem of uneven brightness.
[0021] Taking advantage of the characteristics of high brightness concentration and regular rectangular outline of digital tube areas, edge detection is used in combination with morphological operations (closing operation to connect neighboring areas) and outline search to select outlines that meet preset conditions such as aspect ratio, area range, and density as candidate digital tube display areas.
[0022] The candidate digital tube display area is corrected (affine transformation / perspective transformation) to correct perspective distortion and obtain a standard rectangular digital tube display area.
[0023] In some preferred embodiments, the digital tube display area is segmented into single characters, see reference. Figure 2 Specifically, it includes: Vertical projection is performed within the digital tube display area, the trough positions of the projection histogram are analyzed, and the troughs are used as the dividing points between characters; Based on the segmentation points and prior knowledge, the digital tube display area is segmented into single characters to obtain at least one single-character image.
[0024] Compared to character segmentation based on reinforcement learning, the single-character segmentation operation described above avoids the need for training with a large amount of labeled data.
[0025] In this embodiment, considering the characteristic that digital tube characters are usually of equal width and uniform spacing, prior knowledge (such as the number of characters) is used to assist in segmentation, and possible problems of adhesion (no obvious troughs in the projection) or oversegmentation are handled (using character width constraints).
[0026] In some preferred embodiments, the stroke feature vector of the single-character image is extracted based on the brightness distribution, specifically including: Based on the LED luminous characteristics (brightest at the center of the stroke, with brightness decreasing towards the edge), the stroke features of the single character image are extracted through local adaptive threshold segmentation. Connected component analysis is performed on the stroke features of the single character image, and one or more features, including area, aspect ratio, direction, and center position, are calculated for each connected component to form the stroke feature vector of the single character image. The stroke feature vector is then normalized to the maximum value to eliminate the influence of overall brightness variation on the feature vector. or, Based on the brightness statistics in the grayscale image, the stroke feature vector of the single character image is extracted by partition statistics and / or grayscale projection. The brightness statistics are directly used, avoiding the error introduced by the binarization process, and the robustness to changes in lighting, slight blurring, uneven brightness and other conditions is significantly improved.
[0027] In this embodiment, the number and attributes of the effective connected domains directly correspond to the on / off state of the digital tube segment.
[0028] In some preferred embodiments, the single-character image is locally adaptively segmented using the Sauvola algorithm, which dynamically adjusts the threshold based on the mean and standard deviation of the pixel neighborhood, thereby better preserving weak edges and addressing uneven lighting.
[0029] In some preferred embodiments, the stroke feature vectors are identified based on preset matching rules, see below. Figure 2 Specifically, it includes: For different types of digital tubes (7-segment, 14-segment, 16-segment, etc.), the corresponding matching rules are preset. The matching rules include setting a corresponding ideal stroke feature vector for each single character result (0-9, AF, etc.). Determine the type of digital tube in the original image and select the corresponding matching rule; The similarity between the stroke feature vector and each ideal stroke feature vector is calculated. The stroke feature vector is matched with the ideal stroke feature vector with the highest similarity. Based on the corresponding matching rules, at least one single character result corresponding to a single character image is obtained.
[0030] In this embodiment, the ideal stroke feature vector is used to determine which segments of the digital tube display area are lit and which are off.
[0031] The similarity between the stroke feature vector and each ideal stroke feature vector is calculated using cosine similarity, Euclidean distance, etc.
[0032] The key data structures in the matching rules are as follows: # Define the fields corresponding to each number and character. DIGITS_LOOKUP = { (1, 1, 1, 0, 1, 1, 1): 0, (0, 0, 1, 0, 0, 1, 0): 1, (1, 0, 1, 1, 1, 1, 0): 2, (1, 0, 1, 1, 0, 1, 1): 3, (0, 1, 1, 1, 0, 1, 0): 4, (1, 1, 0, 1, 0, 1, 1): 5, (1, 1, 0, 1, 1, 1, 1): 6, (1, 0, 1, 0, 0, 1, 0): 7, (1, 1, 1, 1, 1, 1, 1): 8, (1, 1, 1, 1, 0, 1, 1): 9, (1, 1, 1, 1, 1, 1, 0): 'A', (0, 1, 0, 1, 1, 1, 1): 'B', (1, 1, 0, 0, 1, 0, 1): 'C', (0, 0, 1, 1, 1, 1, 1): 'D', (1, 1, 0, 1, 1, 0, 1): 'E', (1, 1, 0, 1, 1, 0, 0): 'F', } In some preferred embodiments, during the process of matching the stroke feature vector with the ideal stroke feature vector with the highest similarity, a fault tolerance threshold is introduced to allow the feature value fluctuation of the stroke feature vector within a certain range, in order to deal with minor breaks, adhesion, or noise.
[0033] In some preferred embodiments, after obtaining the character recognition result of the digital tube display area, the character recognition result is verified through a verification mechanism to improve the reliability of the final result; The verification mechanism includes any one or more of the following: check bit verification and numerical range check.
[0034] In the specific implementation process, the winding temperature data of the temperature acquisition device built into the transformer in the gymnasium is collected. The images captured by the camera on the Raspberry Pi are then processed by this set of brightness feature analysis algorithms to quickly and accurately identify the current reading of the digital tube LED display. This reduces the need to install temperature sensors on the transformer or develop a communication protocol that needs to be integrated with the current equipment, greatly improving work efficiency and reducing maintenance costs.
[0035] Example 2 This embodiment further provides a digital tube character recognition system based on Embodiment 1, utilizing a digital tube character recognition method as described above. (See reference...) Figure 3 ,include: The image acquisition module is used to acquire the original image containing the digital tube display area; The image preprocessing module is used to preprocess the original image and filter out the digital tube display areas that meet the preset conditions. A character segmentation module is used to segment the digital tube display area into single characters to obtain at least one single character image; The feature extraction module is used to extract the stroke feature vector of the single character image based on the brightness distribution; The character recognition module is used to recognize the stroke feature vector based on a preset matching rule to obtain a single character result corresponding to at least one single character image; The character combination module is used to combine all the recognized single character results to obtain the character recognition results of the digital tube display area.
[0036] This system analyzes brightness characteristics, making it highly targeted and accurate. It employs a Retinex-based theoretical illumination compensation algorithm to initially mitigate the effects of non-uniform illumination. It does not require high-performance computing units or large-capacity storage, allowing it to be embedded into small industrial PCs or core boards. It eliminates manual copying and data entry, avoiding secondary errors, and automates the entire data acquisition and analysis process, providing a solid data foundation for intelligent manufacturing, quality traceability, and data analysis.
[0037] It is understood that the system in this embodiment corresponds to the method in Embodiment 1 above, and the options in Embodiment 1 above are also applicable to this embodiment, so they will not be described again here.
[0038] Example 3 This embodiment provides a computer-readable storage medium storing at least one instruction, at least one program, code set, or instruction set. The at least one instruction, at least one program, code set, or instruction set is loaded and executed by a processor, causing the processor to perform some or all of the steps of the method provided in Embodiment 1 of this application.
[0039] It is understood that the storage medium can be transient or non-transient. Exemplarily, the storage medium includes, but is not limited to, various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.
[0040] By way of example, the processor may be a central processing unit (CPU), a microprocessor unit (MPU), a digital signal processor (DSP), or a field programmable gate array (FPGA), etc.
[0041] By way of example, the read-only memory includes, but is not limited to, MASK ROM, PROM, EPROM, EEPROM, Flash, etc.
[0042] By way of example, the random access memory includes, but is not limited to, DRAM, SRAM, SDRAM, DDR SDRAM, etc.
[0043] In some examples, a computer program product is provided, which can be implemented by hardware, software, or a combination thereof. As a non-limiting example, the computer program product can be embodied in the storage medium, or it can be embodied in a software product, such as an SDK (Software Development Kit).
[0044] As a non-limiting example, a computer program product is provided, comprising a computer program or computer-executable instructions stored in a computer-readable storage medium. A processor of an electronic device reads the computer program or computer-executable instructions from the computer-readable storage medium, and executes the computer-executable instructions, causing the electronic device to perform some or all of the steps of the method described in the embodiments of this application.
[0045] In some examples, a computer program is provided, including computer-readable code, wherein, when the computer-readable code is run in a computer device, a processor in the computer device performs some or all of the steps for implementing the method.
[0046] This embodiment also proposes an electronic device, including a memory and a processor. The memory stores at least one instruction, at least one program, code set, or instruction set. When the processor executes the at least one instruction, at least one program, code set, or instruction set, it implements some or all of the steps of the method described in Embodiment 1.
[0047] In some examples, a hardware entity of the electronic device is provided, including: a processor, a memory, and a communication interface; wherein the processor typically controls the overall operation of the electronic device; the communication interface is used to enable the electronic device to communicate with other terminals or servers via a network; the memory is configured to store instructions and applications executable by the processor, and may also cache data to be processed or already processed (including but not limited to image data, audio data, voice communication data, and video communication data) to be processed by the processor and various modules in the electronic device, and may be implemented using flash memory (FLASH), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or random access memory (RAM).
[0048] A processor may include one or more processing elements. Therefore, a processor may include one or more integrated circuits (ICs) configured to perform the functions of the processor. Furthermore, each integrated circuit may include circuitry (e.g., a first circuit, a second circuit, and other circuitry) configured to perform the functions of the processor.
[0049] Furthermore, data can be transferred between the processor, communication interface, and memory via a bus, which can include any number of interconnected buses and bridges, connecting various circuits of one or more processors and memories together.
[0050] It is understood that the options in Embodiment 1 above also apply to this embodiment, so they will not be described again here.
[0051] The same or similar labels correspond to the same or similar parts; The terms used to describe positional relationships in the accompanying drawings are for illustrative purposes only and should not be construed as limiting this application. It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other.
[0052] In different specific implementations, the methods or systems described in this application can be implemented in software, hardware, or a combination thereof. Furthermore, the order of the method steps can be changed, and various elements can be added, reordered, combined, omitted, or modified.
[0053] Obviously, the above embodiments of this application are merely examples for clearly illustrating this application, and are not intended to limit the implementation of this application, nor are they intended to limit this application. For those skilled in the art, other variations or modifications can be made based on the above description. The separate structural / functional modules or units can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part. The structure and function of the separate components can be implemented as a combined structure or component. It is neither necessary nor possible to exhaustively describe all embodiments here. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of the claims of this application.
Claims
1. A method for recognizing characters on a digital tube, characterized in that, include: Obtain the original image containing the digital tube display area; The original image is preprocessed to select digital tube display areas that meet preset conditions; The digital tube display area is segmented into single characters to obtain at least one single character image; Stroke feature vectors of the single character image are extracted based on brightness distribution; The stroke feature vector is identified based on a preset matching rule to obtain at least one single character result corresponding to a single character image. All the identified single character results are combined to obtain the character recognition results of the digital tube display area.
2. The digital tube character recognition method according to claim 1, characterized in that, The original image is preprocessed, specifically including: The original image is adaptively illuminated using a Retinex-based theoretical illumination compensation method to obtain a grayscale image. Edge detection, morphological operations, and contour finding are used to locate the ROI region of the grayscale image and filter out candidate digital tube display areas that meet preset conditions; The candidate digital tube display area is corrected to obtain a digital tube display area that meets the preset conditions; The preset conditions include any one or more of aspect ratio, area range, and density.
3. The method for recognizing characters on a digital tube according to claim 1, characterized in that, The digital tube display area is divided into single-character segments, specifically including: Vertical projection is performed within the digital tube display area, the trough positions of the projection histogram are analyzed, and the troughs are used as the dividing points between characters; Based on the segmentation points and prior knowledge, the digital tube display area is segmented into single characters to obtain at least one single-character image.
4. The digital tube character recognition method according to claim 2, characterized in that, The stroke feature vector of the single character image is extracted based on the brightness distribution, specifically including: Based on the light-emitting characteristics of LEDs, the stroke features of the single character image are extracted by local adaptive threshold segmentation; connected component analysis is performed on the stroke features of the single character image, and any one or more features of each connected component, including area, aspect ratio, direction and center position, are calculated to form the stroke feature vector of the single character image, and the stroke feature vector is normalized according to the maximum value. or, Based on the brightness statistics in the grayscale image, the stroke feature vector of the single character image is extracted through partition statistics and / or grayscale projection.
5. The digital tube character recognition method according to claim 4, characterized in that, The single-character image is segmented locally using the Sauvola algorithm, which dynamically adjusts the threshold based on the mean and standard deviation of the pixel neighborhood.
6. The digital tube character recognition method according to claim 4, characterized in that, The stroke feature vector is identified based on a preset matching rule, specifically including: For different types of digital tubes, corresponding matching rules are preset, and the matching rules include setting a corresponding ideal stroke feature vector for each single character result; Determine the type of digital tube in the original image and select the corresponding matching rule; The similarity between the stroke feature vector and each ideal stroke feature vector is calculated. The stroke feature vector is matched with the ideal stroke feature vector with the highest similarity. Based on the corresponding matching rules, at least one single character result corresponding to a single character image is obtained.
7. The digital tube character recognition method according to claim 6, characterized in that, In the process of matching the stroke feature vector with the ideal stroke feature vector with the highest similarity, a fault tolerance threshold is introduced to allow the feature value fluctuation of the stroke feature vector within a certain range.
8. A method for recognizing characters on a digital tube according to any one of claims 1-7, characterized in that, After obtaining the character recognition result of the digital tube display area, the character recognition result is verified through a verification mechanism; The verification mechanism includes any one or more of the following: check bit verification and numerical range check.
9. A digital tube character recognition system, utilizing a digital tube character recognition method as described in any one of claims 1-8, characterized in that, include: The image acquisition module is used to acquire the original image containing the digital tube display area; The image preprocessing module is used to preprocess the original image and filter out the digital tube display areas that meet the preset conditions. A character segmentation module is used to segment the digital tube display area into single characters to obtain at least one single character image; The feature extraction module is used to extract the stroke feature vector of the single character image based on the brightness distribution; The character recognition module is used to recognize the stroke feature vector based on a preset matching rule to obtain a single character result corresponding to at least one single character image; The character combination module is used to combine all the recognized single character results to obtain the character recognition results of the digital tube display area.
10. A computer-readable storage medium, characterized in that, The storage medium stores at least one instruction, at least one program, code set, or instruction set, wherein the at least one instruction, at least one program, code set, or instruction set is loaded and executed by a processor to implement the method as described in any one of claims 1-8.