CN1848136A - Method and system for decoding barcode images - Google Patents

Abstract

The present invention relates to a method for decoding bar code image and its system. Said invention adopts a redundant decode procedure related to data area and text area of bar code, and said method includes the following steps: receiving previewing image of bar code by image sensor of camera, in which the previewing image includes data area representing coded data and text area representing text character; displaying picture of previewing image in previewing window of camera, in which the picture includes pictures of data area and text area; then shooting data image including at least a portion of data area and text area, analyzing data area of data image and identifying optical character of text area of data image so as to decode said bar code.

Description

Method and system for decoding barcode images

technical field

The present invention relates generally to decoding encoded data, and more particularly to methods and systems for decoding images of barcode data, including text data, taken with a digital camera.

Background technique

Barcode scanners are now commonly used in supermarkets, retail stores, and a variety of locations that involve catalog and inventory management, such as warehouses and factories. Typically, barcodes are affixed to items in catalogs on labels that can be read by optical scanners. Common linear or one-dimensional barcodes, such as those following the Uniform Product Code (UPC) developed by IBM in the 1970s, can be lighted by a "flying spot" laser The appropriate location to read. Typically these scanners are either affixed to inventory management equipment such as conveyor belts at supermarket checkouts, or as hand-held wired or wireless devices.

Two-dimensional (2D) barcodes are also becoming more common. Linear barcodes are usually limited to between 10 and 20 characters and thus serve only as simple references to records in databases (eg part numbers, customer account numbers, etc.). However, 2D barcodes can accomplish the same function with much less space. Likewise, 2D barcodes such as Aztec Code, PDF417, Data Matrix and MaxiCode can also serve as databases themselves, as they encode thousands of characters of machine-readable data. Therefore, the 2D barcode can be used as a portable database or portable data file, which is bound to the product throughout the manufacturing and sales chain. In addition, error correction rules can be embedded in 2D barcodes so that no readable information is lost if the surface area of the code is damaged. Therefore, the widespread use of barcodes and improvements in design will allow barcodes to continue to be popular in the future.

Digital cameras represent another widely adopted technology. The cost and size of digital cameras have decreased rapidly in recent years, while camera quality and picture resolution have continued to improve. Digital cameras are now often integrated with a variety of electronic devices, such as mobile phones, personal digital assistants, and notebook computers.

The prior art includes the revelation of combining digital camera technology with barcode scanners. Such a combination can analyze and decode a digital image of a barcode, rather than a laser scan of the barcode. Barcode scanners can thus be used not only by retailers and catalog managers, but also by consumers who scan product barcodes to find more information about that product or store product prices and other information for future reference. However, if the barcode is not decoded immediately, then the digital image of the barcode presents certain problems, because there is no way to confirm whether the digital image of the barcode is of sufficient quality to be successfully decoded. Handheld laser barcode scanners typically emit a familiar beep when the system successfully acquires and decodes a barcode. Especially when scanning partially denuded or damaged barcodes, the user of the scanner often needs to repeatedly manipulate the position of the barcode relative to the scanner until the scanner signals a successful read of the barcode. On the other hand, a user of a digital camera may capture an image of a barcode that cannot be decoded immediately. At this point, low-quality images such as barcodes that are blurry, out of focus, or of insufficient resolution or size may not be read, and the user may no longer have access to the barcode to acquire a new, high-quality image .

Contents of the invention

According to one aspect, the invention is a method of decoding a barcode image. The method includes receiving, at an image sensor of the camera, a preview image of the barcode, wherein the preview image includes a data region representing encoded data and a text region representing text characters. Then display the picture of the preview image in the preview window of the camera, wherein the picture includes the picture of the data area and the text area. A data image including at least a portion of the data region and the text region is then captured. The barcode is decoded by analyzing the data area of the data image and performing optical character recognition on the text area of the data image.

According to another aspect, the invention is a camera system for decoding barcode images. The system includes a camera frame and an image sensor coupled to the camera frame adapted to receive a preview image of a barcode. The preview image includes a data area representing encoded data and a text area representing text characters. A preview window connected to the camera frame is adapted to display a frame of a preview image, and assists in taking a data image including at least a part of the data area and the text area. The processor completes the decoding of the barcode by analyzing the data area of the data image and performing optical character recognition on the text area of the data image.

Description of drawings

For a better understanding of the invention and its practical effects, reference is made to exemplary embodiments described with reference to the accompanying drawings, wherein like reference numerals indicate like elements, wherein:

1 is a schematic diagram illustrating a front face of a camera system according to an embodiment of the present invention;

2 is a schematic diagram illustrating a rear face of a camera system according to an embodiment of the present invention;

Figure 3 is a schematic illustration of a typical, partially blurred digital image of a barcode taken with a prior art camera.

FIG. 4 is a schematic diagram of a barcode image displayed in a preview window of a camera system according to an embodiment of the present invention.

FIG. 5 is an illustration of a cropped barcode image that may be captured, stored, and decoded based on the barcode image shown in FIG. 4 .

FIG. 6 is a flowchart illustrating method steps in accordance with an embodiment of the invention.

Detailed ways

Referring to FIG. 1 , it is a schematic diagram illustrating a camera system 100 according to an embodiment of the present invention. Like most cameras, the system 100 includes a camera housing 105 that integrates a front work surface 110 . The front working surface 110 includes an image sensor 115 adapted to receive images. Image sensor 115 may include conventional lenses and optics as well as digital image sensors such as charge-coupled device (CCD) sensors, CMOS sensors, and the like.

Referring to FIG. 2 , it is a schematic diagram of the rear working surface 120 of the camera system 100 . Rear work surface 120 includes a preview window 125 adapted to display image frames received by image sensor 115 . Similar to most digital cameras, those skilled in the art know that the preview window 125 can be used as a viewfinder of a conventional camera and display any moving image of the scene in front of the image sensor 115 in real time. Typically, the live image displayed by preview window 125 exists only briefly and is stored in memory of system 100 when a user activates a shutter button (not shown), a timer, or other triggering device for image capture. One embodiment of the present invention further includes a mode selector switch 130 for setting the camera system 100 into different photographing modes. For example, like most prior art devices, the mode selector switch 130 can be used to set the system 100 into a standard still image mode, a video mode, a low light mode, a high speed mode, and the like. Mode selector switch 130 may also be used to place system 100 in a barcode reading mode in accordance with the present invention. When the mode selector switch 130 is set to the barcode reading mode, a barcode data capture indicator 135 is displayed on the preview window 125 . Those skilled in the art will appreciate that the mode selector switch 130 may be a physical switch or a toggle switch on the camera housing 105, or a soft switch activated using the preview window 125, for example. The interior of the chassis 105 also includes a processor 140 coupled to the image sensor 115 preview window 125 and mode selector switch 130 to help control the interaction between those components for processing image data.

Different embodiments of the data capture indicator 135 may include different sizes, shapes and formats, and may be translucent or opaque. The in-use data capture indicator 135 is used to guide a user of the system 100 to properly position the image sensor 115 relative to encoded data, such as a barcode. Accordingly, the position of the in-use data capture indicator 135 will be fixed relative to the preview window 125 . According to one embodiment, the data capture indicator 135 is designed to simulate a conventional barcode laser scanning raster pattern, so that it is displayed as a flashing red line near the center of the preview window 124 in the preview window 125 .

As shown in Figure 2, in the operation of an embodiment of the present invention, the user places the data shooting indicator 135 horizontally on the linear barcode, so that the left and right edges of the data shooting indicator 135 just extend to the first barcode respectively. and the last barcode. In addition, the user vertically places the data capture indicator 135 so that the indicator is just positioned on the corresponding text data at the lower end of the barcode. In this position, the system 100 is designed to capture an image of the barcode and its corresponding text data that is in focus, clear, and of sufficient resolution. This image can be efficiently and reliably analyzed and the barcode decoded using decoding algorithms. Also, by taking a clear image of the corresponding text data, optical character recognition (OCR) technology can be reliably applied to the corresponding text data. OCR parses the corresponding text data, and then compares it with the decoded barcode data to confirm the correctness of the decoding process.

Because the data capture indicator 135 can mimic the raster of a conventional barcode laser scanner, users of the barcode reading mode of embodiments of the system 100 can generally use the data capture indicator 135 very easily and intuitively. That's because most people are familiar with laser scanners when shopping in retail stores.

Referring to FIG. 3 , there is an illustration of a typical, partially blurred digital image of a barcode including a data field 305 and a text field 310 taken with a prior art camera. Although the image shown in FIG. 3 is partially blurred and part of the barcode in the data area 305 of the barcode cannot be decoded, the characters in the text area 310 of the barcode are still readable. Mainly due to the relative size and distance between the barcode and the characters in the barcode containing character data, it is common that the data area 305 cannot be decoded while the text area 310 of the barcode is readable. Text characters are generally not as easily blurred as barcode data. In addition, the text characters corresponding to barcodes are usually spaced far apart, making it easier for an OCR program to segment and recognize them. As we all know, most errors in OCR programs are caused by errors in segmenting characters. The present invention takes advantage of the above facts to provide an improved method and system for decoding barcode images.

Referring to FIG. 4 , it is an illustration of displaying a barcode image in the preview window 125 of the camera 100 according to an embodiment of the present invention. The barcode reading mode of the present invention is activated in FIG. 4, so that the preview window 125 includes a data capture indicator 135, which is appropriately displayed at the position of all barcodes passing through the barcode data area 305, and in the text area 310 of the barcode. overlay. This proper location helps to capture an accurately decodable image, where all decodable elements of the barcode, such as all lines of a linear barcode, are captured with sufficient clarity and the corresponding text data can be reliably recognized by the OCR program.

Referring to FIG. 5 , there is an illustration of a cropped barcode image that can be captured, stored, and decoded based on the barcode image shown in FIG. 4 . The image of this cropped area has a number of advantages for the present system 100 over those of prior art bar code readers associated with digital images. First, the image shown in Figure 5 is very small and compact, thus requiring less storage space than the full-size image of the barcode shown in Figure 4; second, the image shown in Figure 5 only captures the relevant data part. This means that the decoding algorithm can work directly on the image without any pre-positioning, registration, positioning or cropping steps. Therefore, the decoding algorithm can be simplified, which is very beneficial in handheld devices with limited storage and computing resources. According to various embodiments of the invention, the size of the captured image area may be optimized for efficient decoding by specific image analysis and decoding algorithms. The size of the captured image area as shown in FIG. 5 can therefore be adjusted by the user or preset by the manufacturer.

Those skilled in the art know that digital camera miniaturization techniques allow the camera frame 105, image sensor 115, and preview window 125 to be of almost any size, so that embodiments of the present invention can be integrated into devices such as mobile phones, personal digital assistants, and notebook computers. in various electronic devices. Furthermore, the data capture indicator 135 can be designed for various types of encoded data, not just linear barcode data. For example a circular or oval shaped data capture indicator 135 may be adapted to aid in reading various QR codes such as Aztec Code, PDF417, DataMatrix and MaxiCode which sometimes also include corresponding text characters.

Referring to FIG. 6 , it is a flowchart illustrating steps of a method 600 according to an embodiment of the present invention. First, at step 605 , a preview image is received at the image sensor 115 of the camera system 100 . The preview image includes a data area 305 representing encoded data such as a barcode and a text area 310 representing text characters. As mentioned above, the preview image is generally a temporary image used to align the camera system 100 to the encoded data. Next, at step 610, the preview window 125 associated with the camera system 100 displays a frame of the preview image. The preview window 125 may include a display of a data capture indicator 135 superimposed on the preview image screen. In step 615, a data image including at least a part of the data area 305 and the text area 310 in the preview image is captured.

Next, at step 620 of method 600 , the data region 305 is analyzed and the noise level of the data region 305 is estimated. Estimation of the noise level is an assessment of the quality of the captured data image and can be accomplished using various techniques well known to those skilled in the art. For example, a severely blurred data image usually has a high noise level, while a sharp and clear data image has a low noise level. At step 625, a decoding algorithm associated with the camera system 100 is used to determine whether the data region 305 is decodable. If the data region 305 is determined not to be decodable, then the method 600 performs an OCR process on the text region 310 of the data image at step 630 . Finally, the result of the OCR process is provided at step 630 as the final decoding result.

The present invention can use various types of OCR programs and systems suitable for downsizing electronic devices. As is well known to those skilled in the art, these OCR systems may include matrix matching, feature extraction, and other types of OCR techniques.

Going back to step 625 of method 600 , if the data area 305 is determined to be decodable, then method 600 uses any barcode decoding algorithm known to those skilled in the art to decode the data area 305 at step 640 . In step 645 it is determined whether the data area 305 was correctly decoded. Many linear and 2D barcodes include error detection mechanisms to confirm correct decoding results. Therefore, if the data region 305 is determined to be correctly decoded, the method 600 directly proceeds to step 635, and the decoded data region 305 is used as the final decoding result.

Alternatively, if it is determined at step 645 that the data region 305 was not decoded correctly, then at step 650 of the method 600 an OCR process is performed on the text region 310 . At step 655 , the results of the decoded data region 305 are compared with the results of the OCR processing of the text region 310 . Next, at step 660 , an OCR confidence score is determined for each character of the data image text area 310 . At step 665 , based on the comparison result of step 655 and the confidence score of step 660 , the particular character identified at step 630 is replaced with the corresponding decoded character obtained at step 640 . Such replacement is typically performed when the OCR confidence score of the replaced character is below a threshold set by the user or preset by the manufacturer of the camera system 100 . Method 600 then provides the final decoding result at step 635 .

In summary, the present invention is an improved method and system for decoding barcode images. Embodiments of the present invention are user-friendly and intuitive in that they effectively simulate the motion of a hand-held laser scanner, albeit in the entirely different context of the preview window 125 of the camera system 100 . Therefore, the present invention helps the user to accurately and reliably capture decodable images of the encoded data through the correct placement of the camera system 100 relative to the encoded data. Furthermore, captured images of encoded data can be more compact and easier to process by image analysis software than images captured using prior art techniques. The present invention also improves barcode decoding accuracy through redundant decoding of the barcode data area 305 and text area 310 . Decoding text regions processed with standard OCR can sometimes achieve higher decoding accuracy than standard barcode decoding processes, especially when decoding blurred images of barcodes. Finally, the decoding process is optimized by comparing the results of the decoded data region 305 and the recognized text region 310 .

The above detailed description provides typical embodiments only, and is not intended to limit the scope, applicability or structure of the invention. Rather, the detailed description of the exemplary embodiments provides those skilled in the art with an enabling description for implementing exemplary embodiments of the invention. It should be realized that various changes can be made in the function and arrangement of elements and steps without departing from the spirit and scope of the invention as described in the claims.

Claims (12)

1. A method for decoding a barcode image, comprising steps: receiving a preview image of the barcode at the camera's image sensor, the preview image including a data area representing encoded data and a text area representing text characters; Display the picture of the preview image in the preview window of the camera, including the picture of the data area and the text area; taking an image of the data including at least a portion of the data region and the text region; and The barcode is decoded by analyzing the data area of the data image and performing optical character recognition on the text area of the data image. 2. The method of claim 1, wherein analyzing the data region of the data image includes estimating a noise level for the data region of the data image. 3. The method of claim 2, further comprising the steps of: The noise level estimation value determines that the data area of the data image is decodable, and the data area of the data image is decoded to determine the decoded character; comparing the decoded data region results with the OCR results for the text regions of the data image; and Provides the final decoding result. 4. The method of claim 1, wherein displaying the preview image frame comprises superimposing a data capturing indicator on the preview image frame of the preview window. 5. The method of claim 1, wherein the camera is integrated in one of the following devices: a handheld digital camera, a personal digital assistant, a laptop computer, or a mobile phone. 6. The method according to claim 3, after comparing the decoded data area result and the optical character recognition result, further comprising the steps of: determining an optical character recognition confidence score for each character of the text region of the data image; and When the OCR confidence score of the OCR-determined character in the text region is lower than the threshold, the OCR-determined character in the text region is replaced with the decoded character determined in the decoded data region. 7. A camera system for decoding barcode images, the system comprising: camera rack; an image sensor coupled to the camera housing and adapted to receive a preview image of the barcode, the preview image including a data area representing encoded data and a text area representing text characters; a preview window attached to the camera gantry, adapted to display a preview image frame and to facilitate capturing an image of the data comprising at least a portion of the data region and the text region; and A processor for decoding the barcode by analyzing the data region of the data image and performing optical character recognition on the text region of the data image. 8. The system of claim 7, wherein the processor is operable to estimate a noise level of a data region of the data image. 9. The system of claim 8, wherein when the noise level estimate determines that the data region of the data image is decodable, the processor is further configured to decode the data region of the data image to determine a decoded character, and convert the decoded data The region results are compared with the OCR results for the text regions of the data image to provide the final decoding result. 10. The system of claim 7, wherein the preview window includes a data capture indicator superimposed on at least a portion of the data area frame, the data capture indicator being used to assist in capturing a data image including at least a portion of the data area and the text area. 11. The system of claim 7, wherein the system is integrated in one of the following devices: a handheld digital camera, a personal digital assistant, a laptop computer or a mobile phone. 12. The system of claim 9, wherein the processor is further configured to determine an OCR confidence score for each character in the text region of the data image, and to determine the OCR confidence score for the character in the text region Below the threshold, the characters determined by the optical character recognition of the text region are replaced with the decoded characters determined by the decoded data region.

Images