Fig 2 - uploaded by Marco Querini
Content may be subject to copyright.
Source publication
2D color barcodes have been introduced to obtain larger storage capabilities than traditional black and white barcodes. Unfortunately, the data density of color barcodes is substantially limited by the redundancy needed for correcting errors, which are due not only to geometric but also to chromatic distortions introduced by the printing and scanni...
Context in source publication
Context 1
... the best of our knowledge, there is little research in the literature on the color classification for 2D color barcodes. One of the first reported attempt to use color in a 2D barcode can be found in a patent by Han et al. [7], who used reference cells to provide standard colors for correct indexing. Bulan et al. [4] proposed to embed data in two different printer colorant channels via halftone-dot orientation modulation, that is, to print two colors at the same spatial location. This allows to nearly double the capacity of black and white barcodes, which is equivalent to use a 4 -ary color scheme for encoding 2 bit/module . This work was extended in [8] by using three instead of two colorant layers and a interference mitigating design of the orientations of the three colorants to improve capacity. Microsoft HCCB uses a grid of colored triangles to encode data, using a palette of 4 or 8 colors ( 4 -ary color or 8 -ary color scheme). HCC2D, the High Capacity Colored 2-Dimensional Barcode [5] uses a grid of colored squares (using a palette of 4 or 8 colors) and has a symbol structure which builds upon a QR code [9] basis for preserving the QR robustness to distortions. Different color barcode technologies adopt different stategies for classifying colors, that is, for converting analog barcode cells to digital bit streams. For instance, the strategy adopted by the Microsoft HCCB decoder is to make use of a color palette, while Bagherinia and Manduchi [10] proposed an algorithm for decoding barcode elements in a color barcode that does not display its reference colors. Despite the increasing interest in color barcodes, we are not aware of any previous attempt at performing a comparative analysis of the performance of different methods for color classification in III. this 2D framework, C OLOR B which ARCODES is the main contri- bution In this of section, our work. we introduce Experimentation 2D color in barcodes, color classification which take advantage has been previously of colors for addressed achieving in higher other domains data density such than as black color recognition and white barcodes. of objects This in indoor is obtained and outdoor at the price images of coping [11], color with recognition chromatic distortions of license during plates [12] decoding. or skin In order color to detection introduce in the face typical localization decoding and process tracking of [13] a color [14], barcode, where, we similarly describe to our next problem, the HCC2D there code, is the which need will to discriminate be used as a among paradigmatic different example classes of throughout color pixels. the We rest emphasize of the paper. that We results remark from that other most domains of the (e.g., findings pixel reported classification in this into paper skin about color and color non-skin classification color) do for not HCC2D necessarily codes carry apply through to color the classification classification of for color other cells 2D in color barcodes, barcodes because as well. the underlying conditions are rather different. Indeed, color classification in 2D barcodes mainly focuses on classifying color cells with minimal size (e.g., thousands cells per square inch) after undergoing a printing and scanning process, while in other domains color elements have other characteristics (e.g., colors in video frames representing natural images). Furthermore, the effective decoding of color barcodes requires much more accuracy and precision than the other applications considered for color classification. III. 2D C OLOR B ARCODES In this section, we introduce 2D color barcodes, which take advantage of colors for achieving higher data density than black and white barcodes. This is obtained at the price of coping with chromatic distortions during decoding. In order to introduce the typical decoding process of a color barcode, we describe next the HCC2D code, which will be used as a paradigmatic example throughout the rest of the paper. We remark that most of the findings reported in this paper about color classification for HCC2D codes apply to color classification for other 2D color barcodes as well. In this section, we describe our HCC2D code, a 2D color barcode which is made of a matrix of square color cells, whose color is selected from a color palette. Figure 1 illustrates samples of HCC2D codes with 4 and 8 colors. We have designed the HCC2D format with the main goal of increasing the data density while preserving the strong robustness to distortions of Quick Response (QR) codes. QR codes are black and white 2D barcode designed by the Japanese corporation Denso Wave which are quite widespread among 2D barcodes, because their acquisition process appears to be strongly reliable, and are suitable for mobile environments, where this technology is rapidly gaining popularity. Structure of QR codes, and consequently, of HCC2D codes is illustrated in Figure 2, being composed of Function Patterns and Encoding Regions . The Position Detection Patterns , the Alignment Patterns , the Timing Patterns , and the Separators for Position Detection Patterns support the detection process in detecting the presence, the proper orientation and the correct slope of a code into an image. The Format Information describes the error correction level used in the code. As previosly introduced, the higher the correction level, the higher the redundancy and the reliability of the barcode reading process, but the lower the actual data density rate. The Version Information represents the code size, that is, the amount of cells (per side) making up the code. Note that the Version Information alone does not determine the final print out size (expressed in inch 2 or cm 2 ), which also depends on hardware parameters, that is, on the printing resolution and on how many printer dots make up each color cell. Finally the Data and Error Correction Codewords contains data plus redundancy. We designed the HCC2D code preserving all the Function Patterns , the Format Information and the Version Information defined in the QR code. Maintaining the structure and the position of such patterns and critical information allows the HCC2D code to preserve the strong robustness to geometric distortions of QR code. Because the retrieval of the Format Information and of the Version Information is a crucial step during the decoding phase (it may led to reading failures) and its storage requirement is small, there is no significant advantage representing it by color cells. The most important changes are gathered in the Data and Error Correction Codewords area. The most noticeable difference with a QR code is that the modules belonging to the Data and Error Correction Codewords area are of different colors; in a HCC2D code with a palette composed of 4 colors each module is able to encode 2 bit/module , while 3 bit/module are stored using 8 colors. Introducing colors in the Data and Error Correction Codewords area requires to address some issues, which we have described in details in [5]. Consider that during QR code reading only the brightness information is taken into account, while HCC2D codes have to cope with chromatic distortions during the decoding phase. Since the Encoding Region is made of color cells, the HCC2D decoder needs to know the complete color palette in order to decode the symbol. To consider the color palette as an a priori shared knowledge between encoding and decoding processes is not a reliable solution; this is because chromatic distortions would not be properly taken into account, arising differently in each printed and scanned image. The processing should be adaptive to each image for better performance. To make a parallelism with black and white barcodes, QR codes compute an adaptive threshold on each image for discriminating dark and light modules, rather than using static thresholds. In order to ensure adaptation to chromatic distortions arisen in each scanned code, we have introduced in the HCC2D code an additional field, the Color Palette Pattern . This is because color cells of a Color Palette Pattern are supposed to be distorted in the same way color cells of the Encoding Region are. We make use of replicated color palettes either for cluster initialization or for training machine learning classifiers. Figure 3 illustrates Color Palette Patterns in HCC2D codes, located at the boundaries. Note that the Color Palette Patterns are not too close to the three Position Detection Patterns areas and are far away from each other, thus ensuring that they are robust to local distortion. Furthermore, Color Palette Patterns take only 2 rows and 2 columns from a symbol consisting of between 21 and 177 rows and columns, and thus, the overhead is small for high density barcodes. IV. C OLOR C LASSIFIERS FOR 2D C OLOR B ARCODES Since the printing and scanning processes introduce chromatic distortions in color barcodes, the decoding success rate depends on the capacity to correctly classify colors of barcode cells. A barcode cell is correctly classified if its original color (before printing) and the class assigned by the classifier to the cell (after scanning) corresponds to each other. A classifier is an algorithm that distinguishes between a fixed set of classes based on labeled training examples. Algorithms reading black and white barcodes may just use a threshold to separate the two classes (that is, dark and light elements), while cells of color barcodes need to be properly classified in many classes, depending on the number of colors. We distinguish 4 or 8 classes (each representing a reference color) into which color pixels may fall, where each pixel is sampled from a cell of the 2D color barcode to decode. Each class reference color is associated with either a 2-bit sequence or a 3-bit sequence. The sequence length depends on how many bits are modulated into each barcode cell (as ...
Similar publications
The present study’s objective was to ascertain the impact of cognitive intervention strategy on the numerical ability of the students with mathematical disability. Sample consisted of 80 third grade students with 40 in experimental and 40 in control group. Schema based strategy instruction was used to help the students to find, organize, plan and s...
Citations
... There are various types of work that have been done focusing on using the color QR Code technique. We have analyzed a few color QR Code that has been used for the different type of QR Code [38][39][40][41][42]. After investigating those systems, we have identified that there are different models and approaches for different QR Codes. ...
... This will generate a multiplexed single-color QR Code with increased capacity. By analyzing different color QR systems [40][41][42]44], we have identified that there are very few common patterns. Most of the patterns are different from one another and have distinct processes for encoding and decoding. ...
... We also observe that there is a decrease in performance between the BSB and WSB or standard QR Code and Multi-Color QR Code [38,45] which indeed resulted in a focusing point for the proposed technique. There are very few common patterns where most of the patterns are different from one another but many techniques of color QR Code increase the importance of these techniques [38][39][40]44] which indeed is resulted as a selecting key to going on a root level pattern technique for the proposed technique. ...
The usage of Quick Response (QR) codes has become widely popular in recent years, primarily for immense electronic transactions and industry uses, where it becomes much easier for users to use different types of applications using the QR codes. The structural flexibility of QR code architecture opens many more possibilities for researchers in the domain of the Industrial Internet of Things (IIoT). However, the limited storage capacity of the traditional QR codes still fails to stretch the data capacity limits. The researchers of this domain have already introduced different kinds of techniques, including data hiding, multiplexing, data compression, color QR code, and so on. However, the research on increasing the data storage capacity of the QR codes is very limited and still operational. The main objective of this work is to increase the data storage capacity of QR codes in the IIoT domain. In the first and main part of the proposed model, We have introduced a 4-States pattern-based encoding technique to generate the proposed 4-States QR (4SQR) code where actual data are encoded into a 4SQR code image which increases the data storage capacity more than the traditional 2-States QR code. The Proposed 4SQR code consists of four types of patterns, including Black Square Box (BSB), White Square Box (WSB), Triangle, and Circle, whereas the traditional 2-States QR codes consist of BSB and WSB. In the second part, the 4SQR code decoding module has been introduced using the adaptive YOLO V5 algorithm where the proposed 4SQR code image is decoded into the actual data. The proposed model is tested in a digital twin framework using randomly generated 3000 testing samples for the encoding module that converts into 4SQR code images successfully and similarly for the decoding module that decodes the 4SQR code images into the actual data. Moreover, experimental results show that this proposed technique offers increased data storage capacity two times than traditional 2-States QR codes.
... Being parallel to their previous work of Querini and Italiano (2013), the same authors experimentally studied to identify the most suitable color classifiers also in . Different algorithms (minimum-distance, logistic tree, k-means clustering, probabilistic naive Bayes, support vector machines, and additionally community-detection in the later study) were tested. ...
Efficiently transferring information, which is of major importance for almost all engineering systems, requires a data representative appropriate to maintain the density and consistency of the information dispatched. As such a machine-readable representation of data, barcodes are one of the most recognized and powerful instruments of the purpose in every domain. Likewise, from the perspective of engineering, colors are excellent sources of information exchange, since the transmission of color connotes the conveyance of its entire scalar attributes in the same spatial channel component. Exerting colors on barcodes as an effective way of bursting the data conveyance capacity has been an active area of research for over 50 years. Significant progress has been achieved through efforts in this regard. It is also envisagable that the evident evolution in related technologies exoterically empowers the enhancement of color barcode capabilities as capacity and reliability, thereby further encouraging prospected research in this direction. Herein, a comprehensive survey of the studies on this main area of interest is presented. To help better acquainted with the field, also a taxonomy of the peculiar interference sources and distortion effects is provided, besides, the 3D barcoding process itself and the research areas are described. Most of the relevant works from debut to the present are broadly examined. Rather than presenting a timeline, studies that pertain to similar issues are addressed together. Amongst those related, premising or pivotal ones are preferably cited as far as feasible. Moreover, all perused works are analyzed by the research areas and the results are presented. Also, the issues relatively more prominent as affecting the performance of the whole process are specified. In the conclusion, some of the research subjects that appear open, scarce, or require further elaboration were remarked on as well. It is anticipated this study to contribute to the efforts toward leveraging color in barcodes.
... To decode color barcodes from blurry images, H. Bagherinia and R. Manduchi [23] propose an iterative method to address the blur-induced color mixing from neighboring color patches [3]. Other researchers have extended traditional QR codes to color QR codes to increase their data capacity [18] [20] [24]. HCC2D proposed in [18] encodes multiple data bits in each color symbol and adds extra color symbols around the color QR codes to provide reference data in the decoding process. ...
Quick Response (QR) code getting widely popular as mobile computing is increasing. But QR code has data limit problem which researchers tried to solve by making color QR codes. But having color QR codes decoding from printed material has lots of image processing problems which are due to different printers color depth, paper quality, environmental effect, dust, light glare and other environmental effects. In our proposed process, we have applied area-based bit detection by percentage of the color level instead of the general binary color bit detection system. This is independent to the printer and the printed substances effect. There is no needed of color correction palette. We also analyzed the performance of other QR codes with us and we find significant error reductions in our technique. In our process, we used only black and white color to print in each bit module and from the percentage of area covered we deducted the level of color percentage despite positions of color bits in that module which increased its scanning speed and also reduced noise error rate to less than 10 percent
... Other researchers have extended traditional QR codes to color QR codes to increase their data capacity [2] [6] [10]. HCC2D proposed in [2] encodes multiple data bits in each color symbol and adds extra color symbols around the color QR codes to provide reference data in the decoding process. ...
The use of color in QR codes brings extra data capacity, but also inflicts tremendous challenges on the decoding process due to chromatic distortion, cross-channel color interference and illumination variation. Particularly, we further discover a new type of chromatic distortion in high-density color QR codes, cross-module color interference, caused by the high density which also makes the geometric distortion correction more challenging. To address these problems, we propose two approaches, namely, LSVM-CMI and QDA-CMI, which jointly model these different types of chromatic distortion. Extended from SVM and QDA, respectively, both LSVM-CMI and QDA-CMI optimize over a particular objective function to learn a color classifier. Furthermore, a robust geometric transformation method is proposed to accurately correct the geometric distortion for high-density color QR codes. We put forth and implement a framework for high-capacity color QR codes equipped with our methods, called HiQ. To evaluate the performance of HiQ, we collect a challenging large-scale color QR code dataset, CUHK-CQRC, which consists of 5390 high-density color QR code samples. The comparison with the baseline method [2] on CUHK-CQRC shows that HiQ at least outperforms [2] by 188% in decoding success rate and 60% in bit error rate. Our implementation of HiQ in iOS and Android also demonstrates the effectiveness of our framework in real-world applications.
... In this work models based on Gaussian distribution, 1D histograms and 3D histograms were evaluated. More advanced approaches like Gaussian Mixture Models [14], artificial neural networks [15] or typical machine learning [16] approach were not considered. However, if in future versions of the system, more, especially quite similar, colours should be recognized, than this methods could provide better performance and reliability. ...
In the article a vision system for shape and colour recognition of dishes (plates, bowls, mugs), which can be used to automate the process of customer service in a self-service canteen is described. In consists of three basic components: object segmentation using so-called background model subtraction, shape recognition using geometric invariant moments and SVM classifier, as well as colour recognition using a Gaussian model. In addition, recognition in case of close or abut objects using a distance transform like approach is presented. The solution was evaluated on a dedicated test stand with controlled LED lightning. A 98% accuracy was obtained on over 100 test images, which indicates that the solution could be used in business practise.
... creases the storage space by using 4, 8, or 16 reference colors in the palette. Querini and Italiano [7] investigated the decoding error rate on HCC2D color barcodes by using different classifiers such as Minimum Distance, Decision Trees, K-means, Naive Bayes, and Support Vector Machines (SVM) classifiers. They showed that K-means algorithm is the most effective classifier in their experiments. ...
The use of colors increases the information storage capacities in barcodes. Increasing the number of colors to encode information makes the decoding a challenging task due to the dependency of the surface color on the illuminant spectrum, viewing parameters, printing device and material, color fading in addition to other nuisance parameters. A popular solution is the use of a color palette of reference colors printed with barcode. The decoding becomes more challenging if a mobile phone as a decoding device is used due to the capture of images from different distances as well as angles. In addition, the barcode images are often blurry because of incorrect focus or camera shake. We present an iterative decoding algorithm that decodes the colors of all barcode patches across the barcode by minimizing the overall observation error. Our method is able to decode colors in presence of blur using a small number of colors yet ensuring high information density.
... Barcodes are optical machine-readable representations of data, capable of storing digital information. Barcode data are represented as a sequence of bytes, which are then mapped A preliminary version of this paper was presented at the Federated Conference on Computer Science and Information Systems [19]. ...
... The first striking difference with the experiments on scanners is that K-means is no longer the most effective algorithm, and it appears to degrade its performance in case of strongly non-uniform illumination of the barcodes, a problem which occurs frequently with mobile phone cameras. This is somewhat different from our preliminary experiments reported in [19], where the data set considered contained pictures taken under uniform light conditions. To illustrate this phenomenon more in detail, we sorted the barcode images in our new data set by error rates (ByER values), and defined good barcode scans and bad barcode scans respectively as the top and the bottom 25% images in this ranking. ...
2D color barcodes have been introduced to obtain larger storage capabilities than traditional black and white barcodes. Unfortunately, the data density of color barcodes is substantially limited by the redundancy needed for correcting errors, which are due not only to geometric but also to chromatic distortions introduced by the printing and scanning process. The higher the expected error rate, the more redundancy is needed for avoiding failures in barcode reading, and thus, the lower the actual data density. Our work addresses this trade-off between reliability and data density in 2D color barcodes and aims at identifying the most effective algorithms, in terms of byte error rate and computational overhead, for decoding 2D color barcodes. In particular, we perform a thorough experimental study to identify the most suitable color classifiers for converting analog barcode cells to digital bit streams. To accomplish this task, we implemented a prototype capable of decoding 2D color barcodes by using different methods, including clustering algorithms and machine learning classifiers. We show that, even if state-of-art methods for color classification could be successfully used for decoding color barcodes in the desktop scenario, there is an emerging need for new color classification methods in the mobile scenario. In desktop scenarios, our experimental findings show that complex techniques, such as support vector machines, does not seem to pay off, as they do not achieve better accuracy in classifying color barcode cells. The lowest error rates are indeed obtained by means of clustering algorithms and probabilistic classifiers. From the computational viewpoint, classification with clustering seems to be the method of choice. In mobile scenarios, simple and efficient methods (in terms of computational time) such as the Euclidean and the K-means classifiers are not effective (in terms of error rate), while, more complex methods are effective but not efficient. Even if a few color barcode designs have been proposed in recent studies, to the best of our knowledge, there is no previous research that addresses a comparative and experimental analysis of clustering and machine learning methods for color classification in 2D color barcodes.
... For this reason, we designed a color barcode denoted as High Capacity Colored 2-Dimensional (HCC2D) code [12], [13], which is well-suited for this framework because of its high data capacity (if compared with state of art barcodes). Specifically, it is capable of encoding about 4KB/inch 2 (effective data density) with a success rate of 90% (reliability) [14], [15]. We designed a new barcode decoding algorithm based on graph drawing methods [16], which is able to run in few seconds even on mobile devices and to achieve nonetheless high accuracy in the recognition phase. ...
Handwritten Signature Verification (HSV) systems have been introduced to automatically verify the authenticity of a user signature. In offline systems, the handwritten signature (represented as an image) is taken from a scanned document, while in online systems, pen tablets are used to register signature dynamics (e.g., its position, pressure and velocity). In online HSV systems, signatures (including the signature dynamics) may be embedded into digital documents. Unfortunately, during their lifetime documents may be repeatedly printed and scanned (or faxed), and digital to paper conversions may result in loosing the signature dynamics. The main contribution of this work is a new HSV system for document signing and authentication. First, we illustrate how to verify handwritten signatures so that signature dynamics can be processed during verification of every type of document (both paper and digital documents). Secondly, we show how to embed features extracted from handwritten signatures within the documents themselves (by means of 2D barcodes), so that no remote signature database is needed. Thirdly, we propose a method for the verification of signature dynamics which is compatible to a wide range of mobile devices (in terms of computational overhead and verification accuracy) so that no special hardware is needed. We address the trade-off between discrimination capabilities of the system and the storage size of the signature model. Towards this end, we report the results of an experimental evaluation of our system on different handwritten signature datasets.
... In previous work [8], we performed a thorough experimental study of algorithms for decoding color barcodes in desktop scenarios, i.e., when barcodes were printed and scanned on low-cost color laser multifunction printers/scanners. In particular, we evaluated the practical performance of several state-of-theart methods for color classification in this framework. ...
... Note that the last three methods (LMT, Naive Bayes and SVM) require an initial preprocessing phase of supervised learning through training examples. We refer the interested reader to [8] for a description of those methods and of the experiments performed in a desktop environment. We only report here that the main experimental findings of [8] showed that the impact of different color classifiers on the error rate achieved in decoding can be quite significant. ...
... We refer the interested reader to [8] for a description of those methods and of the experiments performed in a desktop environment. We only report here that the main experimental findings of [8] showed that the impact of different color classifiers on the error rate achieved in decoding can be quite significant. In particular, the use of more complicated techniques, such as support vector machines, did not seem to pay off, as they did not achieve better accuracy in classifying color barcode cells. ...
The wide availability of on-board cameras in mobile devices and the increasing demand for higher capacity have recently sparked many new color barcode designs. Unfortunately, color barcodes are much more prone to errors than black and white barcodes, due to the chromatic distortions introduced in the printing and scanning process. This is a severe limitation: the higher the expected error rate, the more redundancy is needed for error correction (in order to avoid failures in barcode reading), and thus the lower the actual capacity achieved. Motivated by this, we design, engineer and experiment algorithms for decoding color barcodes with high accuracy. Besides tackling the general trade-off between error correction and data density, we address challenges that are specific to mobile scenarios and that make the problem much more complicated in practice. In particular, correcting chromatic distortions for barcode pictures taken from phone cameras appears to be a great challenge , since pictures taken from phone cameras present a very large variation in light conditions. We propose a new barcode decoding algorithm based on graph drawing methods, which is able to run in few seconds even on low-end computer architectures and to achieve nonetheless high accuracy in the recognition phase. The main idea of our algorithm is to perform color classification using force-directed graph drawing methods: barcode elements which are very close in color will attract each other, while elements that are very far will repulse each other.
This paper addresses the reading problem of multi-level 2D barcodes over a Print-and-Capture (PC) channel. The prior reading schemes have different limitations to hinder their applications, e.g., suffering from quantization error, being sensitive to the predetermined decision boundaries, and being sensitive to the selection of initial parameters. In this paper, we introduce a machine learning approach to address the above limitations using a new ensemble clustering algorithm. Based on the new ensemble clustering algorithm, we propose two reading schemes of a multi-level 2D barcode. Specifically, the first proposed scheme is named the Ensemble Clustering (EC) reading scheme. In the EC reading scheme, we introduce a weighted ensemble mechanism to assign different weights to different base clustering results. Then, we propose the second scheme, named the Enhanced Ensemble Clustering (EEC) reading scheme, to further improve the reading performance with the help of the reference symbols. We implement our approach and conduct extensive performance comparisons through an actual excremental platform under various multi-level 2D barcodes and various capturing devices. From experimental results, we observe that both proposed reading schemes have better performance than the prior reading schemes. Moreover, the EEC reading scheme has better performance than the EC reading scheme, and their performance gap becomes more apparent as the distortion of a PC channel increases.
