No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Low-computation egocentric barcode detector for the blind
... Up until a decade ago, most of the proposed methods could be classified as classical [3][4][5][6][7][8][9][10][11] or based on machine learning, i.e., the latest trend. ...
... Low-computation egocentric barcode detector for the blind [11] The same authors, a year later, aiming to process frames obtained with a wide-angle video camera, discarded their previous method because it was not resistant to motion blur and came up with a new method consisting of a geometric algorithm to detect parallel lines, where a representative line of the cluster determines the height of the barcode bounding box and an estimator of variations in the bisector of the line accounts for the width of the box. This new method increased by up to 98% the capacity to detect codes on simple datasets and lowered the time cost of processing a frame to 40 ms. ...
... As the YOLO output consists of rectangles containing barcodes, which can appear rotated, they used another network to predict the rotation angle. They obtained the same accuracy as Creusot and Munawar [11], but thanks to the use of a GTX 1080 GPU, they could process a frame in 14 ms. However, this method cannot be considered to have processing time comparable to video acquisition time on CPUs. ...
In this work, two methods are proposed for solving the problem of one-dimensional barcode segmentation in images, with an emphasis on augmented reality (AR) applications. These methods take the partial discrete Radon transform as a building block. The first proposed method uses overlapping tiles for obtaining good angle precision while maintaining good spatial precision. The second one uses an encoder–decoder structure inspired by state-of-the-art convolutional neural networks for segmentation while maintaining a classical processing framework, thus not requiring training. It is shown that the second method’s processing time is lower than the video acquisition time with a 1024 × 1024 input on a CPU, which had not been previously achieved. The accuracy it obtained on datasets widely used by the scientific community was almost on par with that obtained using the most-recent state-of-the-art methods using deep learning. Beyond the challenges of those datasets, the method proposed is particularly well suited to image sequences taken with short exposure and exhibiting motion blur and lens blur, which are expected in a real-world AR scenario. Two implementations of the proposed methods are made available to the scientific community: one for easy prototyping and one optimised for parallel implementation, which can be run on desktop and mobile phone CPUs.
... Over the years, a number of methods have been proposed to detect barcodes using classical signal processing [1,2,3,4,5], but nearly all of them take too long to process Ultra High-Resolution (UHR) images. More specifically, [5] used parallel segment detectors which improved on their previous work [6] of finding imaginary perpendicular lines in Hough space with maximal stable extremal This work was supported by Amazon.com, ...
... Over the years, a number of methods have been proposed to detect barcodes using classical signal processing [1,2,3,4,5], but nearly all of them take too long to process Ultra High-Resolution (UHR) images. More specifically, [5] used parallel segment detectors which improved on their previous work [6] of finding imaginary perpendicular lines in Hough space with maximal stable extremal This work was supported by Amazon.com, Inc. regions to detect barcodes. ...
... Inc is made of 3.8 million UHR images of resolution up to 30k × 30k grayscale images and could not be released due to confidentiality reasons. Additionally, [5] .982 ---.989 ---Hansen et al. [9] .991 ...
Object detection in Ultra High-Resolution (UHR) images has long been a challenging problem in computer vision due to the varying scales of the targeted objects. When it comes to barcode detection, resizing UHR input images to smaller sizes often leads to the loss of pertinent information, while processing them directly is highly inefficient and computationally expensive. In this paper, we propose using semantic segmentation to achieve a fast and accurate detection of barcodes of various scales in UHR images. Our pipeline involves a modified Region Proposal Network (RPN) on images of size greater than 10k$\times$10k and a newly proposed Y-Net segmentation network, followed by a post-processing workflow for fitting a bounding box around each segmented barcode mask. The end-to-end system has a latency of 16 milliseconds, which is $2.5\times$ faster than YOLOv4 and $5.9\times$ faster than Mask RCNN. In terms of accuracy, our method outperforms YOLOv4 and Mask R-CNN by a $mAP$ of 5.5% and 47.1% respectively, on a synthetic dataset. We have made available the generated synthetic barcode dataset and its code at http://www.github.com/viplab/BSBD/.
... Existing approaches to vision-based barcode reading are either geometric-based [2][3][4] or learning-based [5,6]. While geometric approaches perform better in terms of accurateness of segmentation, they tend to be weak in the presence of severe distortion or occlusion. ...
... Meanwhile, for excellent 1D barcode detection results, existing 1D barcode detection approaches need to tune many thresholds set in the algorithms to apply in different detection conditions. The correctness of the threshold value may directly influence the performance of the algorithm, for example, threshold T which determines the candidate for the parallel segment detector by Creusot et al. [3], and the value of threshold ratio k th for bar description by Namane et al. [4]. ...
... Different approaches based on geometry are widely adopted. Creusot et al. [3] find a candidate line segment in the barcode which crops the region in orthogonal direction by intensity value. Namane et al. [4] estimate whether the line segments, transformed from the outer contours, fit the bar description by length and orientation. ...
Vision-based 1D barcode reading has been the subject of extensive research in recent years due to the high demand for automation in various industrial settings. With the aim of detecting the image region of 1D barcodes, existing approaches are both slow and imprecise. Deep-learning-based methods can locate the 1D barcode region fast but lack an adequate and accurate segmentation process; while simple geometric-based techniques perform weakly in terms of localization and take unnecessary computational cost when processing high-resolution images. We propose integrating the deep-learning and geometric approaches with the objective of tackling robust barcode localization in the presence of complicated backgrounds and accurately detecting the barcode within the localized region. Our integrated real-time solution combines the advantages of the two methods. Furthermore, there is no need to manually tune parameters in our approach. Through extensive experimentation on standard benchmarks, we show that our integrated approach outperforms the state-of-the-art methods by at least 5%.
... Cresot's algorithm detects dark bars of barcodes using Maximal Stable Extremal Regions (MSER) followed by finding imaginary perpendicular to bars center line in Hough space. In 2016 Cresot et al. came with a new paper [9] improving previous results using a new variant of Line Segment Detector instead of MSER, which they called Parallel Segment Detector. [10] proposes another bars detection method for 1D barcode detection, which is reported to be absolutely precise in real-time applications. ...
... We compare our results with Cresot2015 [7], Cresot2016 [9], Namane2017 [10], Yolo2017 [11] on Artelab and Muenster datasets (Table II). ...
Universal Barcode Detector via Semantic Segmentation
... (9,10) In any case, the determination of a barcode area is essential for the equipment to recognize the barcodes for analysis. (5,(11)(12)(13)(14) When a common laser-scanning method is employed as the barcode scanner, a barcode area can be located by attaching scanners to each slot into which strips are mounted. However, this requires multiple scanning devices, resulting in the increased size and cost of the overall equipment. ...
... Soros et al [8] continued dealing with blur using structure matrix and saturation from HSV color system to detect blurry barcodes better but with expense of lowing speed in 2013. Recently, Creusot et al. [9] proposed a faster method for blurry barcodes based on Line Segment Detector after their previous work [10] using Maximal Stable Extremal Region shown sensitive to blur. In another way, Hansen [11] first tried to apply an object detection deep learning model (YOLO) on both 1D and 2D codes with the best bounding box detection rate. ...
Barcodes are ubiquitous and have been used in most of critical daily activities for decades. However, most of traditional decoders require well-founded barcode under a relatively standard condition. While wilder conditioned barcodes such as underexposed, occluded, blurry, wrinkled and rotated are commonly captured in reality, those traditional decoders show weakness of recognizing. Several works attempted to solve those challenging barcodes, but many limitations still exist. This work aims to solve the decoding problem using deep convolutional neural network with the possibility of running on portable devices. Firstly, we proposed a special modification of inference based on the feature of having checksum and test-time augmentation, named as Smart Inference (SI) in prediction phase of a trained model. SI considerably boosts accuracy and reduces the false prediction for trained models. Secondly, we have created a large practical evaluation dataset of real captured 1D barcode under various challenging conditions to test our methods vigorously, which is publicly available for other researchers. The experiments' results demonstrated the SI effectiveness with the highest accuracy of 95.85% which outperformed many existing decoders on the evaluation set. Finally, we successfully minimized the best model by knowledge distillation to a shallow model which is shown to have high accuracy (90.85%) with good inference speed of 34.2 ms per image on a real edge device.
... In recent years deep learning and artificial intelligence with Hough transform and morphological operations is increasingly being used to localize and decode barcodes [5,[29][30][31] . The Parallel Line Segment Detector with Hough transform and morphological operations have also been used to decode barcodes in real-time [7,32] . Furthermore, the Zamberletti algorithm has recently been used to detect multiple 1D and 2D barcode images [8] . ...
... In recent years deep learning and artificial intelligence with Hough transform and morphological operations is increasingly being used to localize and decode barcodes [5,[29][30][31] . The Parallel Line Segment Detector with Hough transform and morphological operations have also been used to decode barcodes in real-time [7,32] . Furthermore, the Zamberletti algorithm has recently been used to detect multiple 1D and 2D barcode images [8] . ...
Abstract: Automation of production in the nurseries of flower producing companies using barcode scanners have been
attempted but with little success. Stationary laser barcode scanners which have been used for automation have failed due to
the close proximity between the barcode and the scanner, and factors such as speed, angle of inclination of the barcode, damage
to the barcode and dirt on the barcode. Furthermore, laser barcode scanners are still being used manually in the nurseries
making work laborious and time consuming, thereby leading to reduced productivity. Therefore, an automated image-based
barcode detection system to help solve the aforementioned problems was proposed. Experiments were conducted under
different situations with clean and artificially soiled Code 128 barcodes in both the laboratory and under real production
conditions in a flower producing company. The images were analyzed with a specific algorithm developed with the software
tool Halcon. Overall the results from the company showed that the image-based system has a future prospect for automation
in the nursery.
Keywords: automation, barcode detection, horticultural production systems, image processing, barcode scanners
DOI: 10.25165/j.ijabe.20191206.4762
... Hough transform is applied to the input image to achieve the angle invariable. In the same year, Creusot et al. [7] used a method called Parallel Segment Detector (PSD) which is based on Line Segment Detector (LSD). They used Maximal Stable Extremal to detect the bar code's black stripes, and then use the Hough Transfom to find the vertical line of the barcode. ...
Repetitive tasks widely exist in applied fields of human-computer interaction. One underestimated example is parcel scanning, which has consistent operation difficulty but comprises multiple processes (e.g., label seeking and scanning, result confirming, and parcel relocating), involving respective cognitive requirements. Many devices are developed to facilitate repetitive operations, but few are to reduce fluctuating cognitive load throughout task processes. We present the eye-tracking augmented reality headset that integrates foveated vision detection and smooth pursuit of eye tracking and investigate how it can reduce cognitive load in the repetitive task. In total, 33 participants completed a set of parcel scanning tasks with the headset and their visual and cognitive performance were assessed. The results show that the headset maintained high scanning efficiency and lower cognitive load across the tasks with varying difficulties and it significantly reduced the participants’ cognitive load during the processes of barcode seeking and scanning and result confirmation. The headset demonstrated good usability and ease of use. Implications for how the case study result could be used in generalizing applications are discussed.
Barcode detection is a key step before decoding so that achieving a fast and accurate detection algorithm is of significant importance. In the present study, we propose to guide the pruning of channels and shortcut layers in YOLOv4 through sparse training to obtain the compressed model ThinYOLOv4 for barcode detection. Then a binary classification network is established to remove the prediction boxes that do not contain a barcode, thereby obtaining a fast and accurate barcode detection model. In order to evaluate the performance of the proposed method, a barcode dataset consisting of 16,545 images is provided. This dataset contains common types of barcodes in the market and covers different practical scenarios. Furthermore, interference factors such as blur, low-contrast are considered in the dataset purposefully. Obtained results show that the proposed method achieves a recall rate of 93.8% on the provided dataset, Meanwhile, parameters of YOLOv4 are reduced from 63,943,071 to 400,649, and the model size is reduced from 250,037 KB to 1,587 KB, while the corresponding detection speed is increased to 260% of YOLOv4. When the experiment is performed on the 1050Ti GPU, a detection speed of 23.308 ms/image is achieved.
Given the wide use of barcodes, there is a growing demand for their efficient detection and recognition. However, the existing publicly available datasets are insufficient and of poor quality. Moreover, recently proposed approaches were trained on different private datasets, which makes the comparison of proposed methods even more unfair. In this paper, we propose a simple yet efficient technique to generate realistic datasets for barcode detection problem. Using the proposed method, we synthesized a dataset of \(\sim \)30,000 barcodes that closely resembles real barcode data distribution in terms of size, location, and number of barcodes on an image. The dataset contains a large number of different barcode types (Code128, EAN13, DataMatrix, Aztec, QR, and many more). We also provide a new real test dataset of 921 images, containing both document scans and in-the-wild photos, which is much more challenging and diverse compared to existing benchmarks. These new datasets allow a fairer comparison of existing barcode detection approaches. We benchmarked several deep learning techniques on our datasets and discuss the results. Our code and datasets are available at https://github.com/abbyy/barcode_detection_benchmark.
This paper presents a novel approach for accurate barcodes detection in real and challenging environments using compact deep neural networks. Our approach is based on Convolutional Neural Network (
CNN
) and neural network compression, which can detect the four vertexes coordinates of a barcode accurately and quickly. Our approach consists of four stages: (
$i$
) feature extraction by a base network, (
$ii$
) region proposal network (
RPN
) training, (
$iii$
) barcode classification and coordinates regression, and (
$iv$
) weights pruning and recoding. The model is trained in the first three stages and then compressed in the fourth stage to reduce the size of the trained model. In order to remove the effect of geometric distortion during barcode decoding, we add a distortion removal module to the end of the trained model. In experiments, we validate our approach on a challenging large-scale dataset. Compared with previous methods, our method can locate the coordinates of a barcode accurately and quickly and enhance decoding rate through distortion removal. In addition, the storage and memory overheads of our model are reduced through model compression, which shows great potential in industrial applications.
This paper presents a novel method for efficient barcodes detection in real and complicated environments using a Convolutional Neural Network (CNN)-based model. The method is developed as a preprocess-module of existing decoders to enhance decoding rates. Our method is trained as an end-to-end model to determine accurate locations of four barcode vertexes. Our method consists of four modules: base net module, region proposals generator, classification and regression module, and distortion removal module. The feature of barcodes extracted from the base net is fed to the next module. Region proposals are generated and selected as Region of Interest (ROI). Then the ROI are forward propagated to the classification and regression module to determine the positions and shapes of the barcodes. Finally, the distortion removal module is used to remove the geometric distortion according to regression parameters acquired from the previous step. The accurate position and distorted barcodes shape can be determined and corrected by our method. We validate our method on a challenging large-scale dataset in experiments. Compared with previous methods, our method provides an end-to-end solution to determine accurate locations of barcode vertexes, which shows an excellent performance on detection accuracy. In addition, our method can enhance decoding rate through distortion removal.
Barcode reading mobile applications that identify products from pictures taken using mobile devices are widely used by customers to perform online price comparisons or to access reviews written by others. Most of the currently available barcode reading approaches focus on decoding degraded barcodes and treat the underlying barcode detection task as a side problem that can be addressed using appropriate object detection methods. However, the majority of modern mobile devices do not meet the minimum working requirements of complex general purpose object detection algorithms and most of the efficient specifically designed barcode detection algorithms require user interaction to work properly. In this paper, we present a novel method for barcode detection in camera captured images based on a supervised machine learning algorithm that identifies one-dimensional barcodes in the two-dimensional Hough Transform space. Our model is angle invariant, requires no user interaction and can be executed on a modern mobile device. It achieves excellent results for two standard one-dimensional barcode datasets: WWU Muenster Barcode Database and ArTe-Lab 1D Medium Barcode Dataset. Moreover, we prove that it is possible to enhance the overall performance of a state-of-the-art barcode reading algorithm by combining it with our detection method.
Barcode technology is essential in automatic identification, and is used in a wide range of real-time applications. Different code types and applications impose special problems, so there is a continuous need for solutions with improved performance. Several methods exist for code localization, that are well characterized by accuracy and speed. Particularly, high-speed processing places need reliable automatic barcode localization, e.g. conveyor belts and automated production, where missed detections cause loss of profit. Our goal is to detect automatically, rapidly and accurately the barcode location with the help of extracted image features. We propose a new algorithm variant, that outperforms in both accuracy and efficiency other detectors found in the literature using similar ideas, and also improves on the detection performance in detecting 2D codes compared to our previous algorithm.
We propose a linear-time line segment detector that gives accurate results, a controlled number of false detections, and requires no parameter tuning. This algorithm is tested and compared to state-of-the-art algorithms on a wide set of natural images.
The linear 1D barcode is the main tagging system for billions of products sold each day. Barcodes have many advantages but require a laser scanner for fast and robust scanning. Solutions exist to read barcodes from cell phones but they assume a carefully framed image within the field of view. This undermines the true potential of barcodes in a wide range of scenarios. In this paper we present a real time technique to detect barcodes in the wild from video streams. Our technique outperforms the state-of-the-art passive techniques both in accuracy and speed. Potential commercial applications enabled by such passive scanning system are also discussed in this paper.
Barcode localization is an essential step of the barcode reading process. For industrial environments, having high-resolution cameras and eventful scenarios, fast and reliable localization is crucial. Images acquired in those setups have limited parameters, however, they vary at each application. In earlier works we have already presented various barcode features to track for localization process. In this paper, we present a novel approach for fast barcode localization using a limited set of pixels in image domain.
The main challenge in developing an image-based bar-code reading system is barcode localization. We propose a non-traditional localization algorithm that initially uses a bot-tom hat filter to highlight dark bars on a light background. The algorithm has been implemented in MATLAB and tested on a variety of realistic and non-ideal images, achieving an overall sensitivity of 98.24% and specificity of 99.19%. In ideal image categories, we are able to achieve 100% accu-racy. The proposed method also shows potential for imple-mentation in real-time applications.
Barcode detection is required in a wide range of real-life applications. Imaging conditions and techniques vary considerably and each application has its own requirements for detection speed and accuracy. In our earlier works we used uniform partitioning with several approaches for detection of various types of 1D and 2D barcodes and showed their behaviour on a set of test images. In this work, we extend the partitioning idea and replace scan-line based methods with distance transformation to improve accuracy.
The use of an expired product may cause harm to its designated target. If the product is for human consumption, e.g. medicine, the result can be fatal. While most people can check the expiration date easily before using the product, it is very difficult for a visually impaired or a totally blind person to do so independently. This paper therefore proposes a solution that helps the visually impaired to identify a product and subsequently 'read' the expiration date on a product using a handheld Smartphone. While there are a few commercial barcode decoder and text recognition applications for the mobile phone, they require the user to point the phone to the correct location - which is extremely hard for the visually impaired. We thus focus our research on helping the blind user to locate the barcode and the expiration date on a product package. After that, existing barcode decoding and OCR algorithms can be utilized to obtain the required information. A field trial with several bind- folded/totally-blind participants is conducted and shows that the proposed solution is effective in guiding a visually impaired user towards the barcode and expiry information, although some issues remain with the reliability of the off-the-shelf decoding algorithms on low-resolution videos.
Barcode technology is the pillar of automatic identification, that is used in a wide range of real-time applications with various types of codes. The different types of codes and applications impose special problems, so there is a continuous need for solutions with improved effectiveness. There are several methods for barcode localization, that are well characterized by accuracy and speed. Particularly, high-speed processing places need automatic barcode localization, e. g. conveyor belts, automated production, where missed detections cause loss of profit. In this paper, we mainly deal with segmentation of images with 1D barcode, but also analyze the operation of different methods for 2D barcode images as well. Our goal is to detect automatically, rapidly and accurately the barcode location by the help of extracted features. We compare some published method from the literature, which basically rely on the contrast between the background and the shape that represent the code. We also propose a novel algorithm, that outperforms the others in both accuracy and efficiency in detecting 1D codes.
In this paper we present an algorithm for the recognition of 1D barcodes using camera phones, which is highly robust regarding the the typical image distortions. We have created a database of barcode images, which covers typical distortions, such as inhomogeneous illumination, reflections, or blurriness due to camera movement. We present results from experiments with over 1,000 images from this database using a Matlab implementation of our algorithm, as well as experiments on the go, where a Symbian C++ implementation running on a camera phone is used to recognize barcodes in daily life situations. The proposed algorithm shows a close to 100% accuracy in real life situations and yields a very good resolution dependent performance on our database, ranging from 90.5% (640 Ã 480) up to 99.2% (2592 Ã 1944). The database is freely available for other researchers.
ZXing Multi-format 1D/2D barcode image processing library
- Zxing