Logo Image Research Articles

Vehicle Logo Recognition with Small Sample Problem in Complex Scene Based on Data Augmentation

Automatic identification for vehicles is an important topic in the field of Intelligent Transportation Systems (ITS), and the vehicle logo is one of the most important characteristics of a vehicle. Therefore, vehicle logo detection and recognition are important research topics. Because of the problems that the area of a vehicle logo is too small to be detected and the dataset is too small to train for complex scenes, considering the speed of recognition and the robustness to complex scenes, we use deep learning methods which are based on data optimization for vehicle logo in complex scenes. We propose three augmentation strategies for vehicle logo data: cross-sliding segmentation method, small frame method, and Gaussian Distribution Segmentation method. For the problem of small sample size, we use cross-sliding segmentation method, which can effectively increase the amount of data without changing the aspect ratio of the original vehicle logo image. To expand the area of the logos in the images, we develop the small frame method which improves the detection results of the small area vehicle logos. In order to enrich the position diversity of vehicle logo in the image, we propose Gaussian Distribution Segmentation method, and the result shows that this method is very effective. The F1 value of our method in the YOLO framework is 0.7765, and the precision is greatly improved to 0.9295. In the Faster R-CNN framework, the F1 value of our method is 0.7799, which is also better than before. The results of experiments show that the above optimization methods can better represent the features of the vehicle logos than the traditional method, and the experimental results have been improved.

Aplikasi Pengenalan Pola Citra Logo Obat Medis Menggunakan K-Means Clustering

Trends in drug abuse have occurred since 2014 such as ATS abuse or chemicals such as over-the-counter PCC. People no longer follow the rules of using drugs according to applicable regulations, but many who buy and consume drugs without a doctor's prescription. These behaviors and actions can endanger health, so people need to know the use of drugs to be more safe for consumption, one of them by paying attention to the type of logo on the drug packaging. To make it easier to find out the name of the logo type on drug packaging, it is necessary to develop an application using digital image processing. The study began with taking a drug pattern image using a smartphone camera and the results were carried out cropping the logo image in detail. Furthermore, cropping (RGB) images are converted into binary images and grayscale images to be performed a feature extraction process that is metric, eccentricity, and GLCM. The results of testing the test image data, a percentage of 91.5% with an accuracy rate of 100% types of over-the-counter drugs, hard drugs, and narcotics, while over-the-counter drug images are limited to 66%.

Sponsors and Facilitators

List of Sponsors and Facilitators logo images available in this PDF.

Design and simulation a video steganography system by using FFT­turbo code methods for copyrights application

Protecting information on various communication media is considered an essential requirement in the present information transmission technology. So, there is a continuous search around different modern techniques that may be used to protect the data from the attackers. Steganography is one of those techniques that can be used to maintain the copyright by employing it to cover the publisher logo image inside the video frames. Nowadays, most of the popular known of the Video-Steganography methods become a conventional technique to the attacker, so there is a requirement for a modern and smart strategy to protect the copyright of the digital video file. Where this proposed system goal to create a hybrid system that combines the properties of Cryptography and Steganography work to protect the copyright hidden data from different attack types with maintaining of characteristics of the original video (quality and resolution). In this article, a modern Video-Steganography method is presented by employing the benefits of TC (Turbo code) to encrypt the pixels of logo image and Least two Significant Bit Technique procedure to embed the encryption pixels inside the frames of the video file. The insertion is performed in the frequency domain by applying the Fast Fourier Transform (FFT)on the video frames. The examination of the suggested architecture is done by terms of Structural Similarity Index, MSE (mean squared error), and PSNR (peak signal-to-noise ratio) by comparing between an original and extracted logo as well as between original and Steganographic video (averaged overall digital frames in the video). The simulation results show that this method proved high security, robustness, capacity and produces a substantial performance enhancement over the present known ways with fewer distortions in the quality of the video

Logo-2K+: A Large-Scale Logo Dataset for Scalable Logo Classification

Logo classification has gained increasing attention for its various applications, such as copyright infringement detection, product recommendation and contextual advertising. Compared with other types of object images, the real-world logo images have larger variety in logo appearance and more complexity in their background. Therefore, recognizing the logo from images is challenging. To support efforts towards scalable logo classification task, we have curated a dataset, Logo-2K+, a new large-scale publicly available real-world logo dataset with 2,341 categories and 167,140 images. Compared with existing popular logo datasets, such as FlickrLogos-32 and LOGO-Net, Logo-2K+ has more comprehensive coverage of logo categories and larger quantity of logo images. Moreover, we propose a Discriminative Region Navigation and Augmentation Network (DRNA-Net), which is capable of discovering more informative logo regions and augmenting these image regions for logo classification. DRNA-Net consists of four sub-networks: the navigator sub-network first selected informative logo-relevant regions guided by the teacher sub-network, which can evaluate its confidence belonging to the ground-truth logo class. The data augmentation sub-network then augments the selected regions via both region cropping and region dropping. Finally, the scrutinizer sub-network fuses features from augmented regions and the whole image for logo classification. Comprehensive experiments on Logo-2K+ and other three existing benchmark datasets demonstrate the effectiveness of proposed method. Logo-2K+ and the proposed strong baseline DRNA-Net are expected to further the development of scalable logo image recognition, and the Logo-2K+ dataset can be found at https://github.com/msn199959/Logo-2k-plus-Dataset.

Organizer Logo

List of Organizer Logo, Sponsor Logo images are available in this pdf.

Logo

List of Logo images are available in the pdf.

Wavelet transform modulus maxima‐based robust logo watermarking

Digital image watermarking is used to protect the copyright of digital images. In this study, a novel blind logo image watermarking technique for RGB images is proposed. The proposed technique exploits the error correction capabilities of the human visual system. It embeds two different watermarks in the wavelet/multiwavelet domains. The two watermarks are embedded in different sub-bands, are orthogonal, and serve different purposes. One is a high capacity multi-bit watermark used to embed the logo, and the other is a 1-bit watermark which is used for the detection and reversal of geometrical attacks. The two watermarks are both embedded using a spread spectrum approach, based on a pseudo-random noise sequence and a unique secret key. Robustness against geometric attacks such as rotation, scaling, and translation (RST) is achieved by embedding the 1-bit watermark in the wavelet transform modulus maximacoefficients of the wavelet transform. The experimental results show that the proposed watermarking technique has better distortion parameter detection capabilities and compares favourably against existing techniques in terms of robustness against geometrical attacks such as RST.

Quantum Image Watermarking Based on Wavelet and Geometric Transformation

Like the digital watermark, which has been highlighted in previous studies, the quantum watermark aims to protect the copyright of any image and to validate its ownership using visible or invisible logos embedded in the cover image. In this paper, we propose a method to include an image logo in a cover image based on quantum fields, where a certain amount of texture is encapsulated to encode the logo image before it is included in the cover image. The method also involves transforming wavelets such as Haar base transformation and geometric transformation. These combination methods achieve a high degree of security and robustness for watermarking technology. The digital results obtained from the experiment show that the values of Peak Signal Noise Ratio (PSNR) and Mean Squared Error (MSE) that apply to one of the set cover images were 68.74 and 0.093, respectively.

A robust zero-watermarking algorithm for color image based on tensor mode expansion

Digital zero-watermarking technology is an effective measure to protect image copyright, and many algorithms have been proposed based on zero-watermark. However, most of the existing zero-watermarking algorithms are designed for grayscale images. In this paper, a new zero-watermarking algorithm for color image based on tensor mode expansion is proposed. In the proposed scheme, four images of R, G, B, and gray are firstly generated from the original image. Then, the four images are appropriately combined to construct two three-dimensional tensors, and tensor expansion is performed on the obtained tensors. By performing singular value decomposition (SVD) and discrete cosine transform (DCT) on the expanded data, a feature image containing the main information of the host image can be generated. Finally, the feature image is fused with the specified logo image to obtain a zero-watermark image. Experimental results show that the proposed algorithm is robust to noise attacks and common image processing attacks, and better performances compared with some existing algorithms are also achieved. In addition, since the proposed algorithm is based on color images, it can make full use of all aspects of color images.

One-Stage Logo Detection Framework using AdaBoost Resnet50 Backbone

Logo is an important asset as it is designed to express identity or character of the company or organization that owns the logo. The advent of deep learning methods and proliferated of logo images sample dataset in the past decade has made automated logo detection from digital images or video an interesting computer vision problem with wide potential applications. This paper presents a novel one-stage logo detector framework in which the backbone of the proposed logo detector is a deep learning model which is trained supervisedly using gradient descent training algorithm and the target logo classes as input dataset. The experiment results showed that AdaBoost Resnet50 (0.58 MAP) as the logo detector backbone outperforms Resnet50 (0.56 MAP), VGG19 (0.32 MAP), and AdaBoost VGG19 (0.56 MAP).

Privacy Protection of Patient Medical Images using Digital Watermarking Technique for E-healthcare System.

Privacy protection has been a critical issue in the delivery of medical images for telemedicine, e-health care and other remote medical systems. The aim of this proposed work is to implement a secure, reversible, digital watermarking technique for the transmission of medical data remotely in health care systems. In this research work, we employed a novel optimized digital watermarking scheme using discrete wavelet transform and singular value decomposition using cuckoo search algorithm based on Lévy flight for embedding watermark into the grayscale medical images of the patient. The performance of our proposed algorithm is evaluated on four different 256 × 256 grayscale host medical images and a 32 × 32 binary logo image. The performance of the proposed scheme in terms of peak signal to noise ratio was remarkably high, with an average of 55.022dB compared to other methods. Experimental results reveal that the proposed method is capable of achieving superior performance compared to some of the state-of-art schemes in terms of robustness, security and high embedding capacity which is required in the field of telemedicine and e-health care system.

A Qualitative Investigation of Academic and Athletic Brand Representations: Applying Lovemarks Theory to Cultural Symbol

Abstract The current study applied Robert’s lovemarks theory of branding to interpret a photo-elicitation approach of two academic/athletic brand representations. Three major themes emerged from the participants (N = 88) in the study of the Thinkman logo image: a) Intellectual Enhancement; b) Athlete; and c) Hard Work. By eliciting responses from the participants (N = 75) regarding the term Scholar-Baller , the following three themes emerged: a) Amazing Scholar & Exceptional Athlete; b) Top Student; and c) Student-Athlete. The current findings indicate many of the participants responded favorably to the photo-elicitation of two academic/athletic terms and logos. The participants also critiqued both representations in the context of the broader and specific discourses about the student-athlete experience in American higher education. Implications and recommendations for scholars and practitioners are suggested.

Smart halal recognizer for muslim consumers

<span>Halal is the term used for permissible food according to Islamic dietary law. Indicators such as Halal logo have been used to facilitate Muslims in identifying Halal food. In Malaysia, the Department of Islamic Development (JAKIM) has introduced a standard Halal logo for locally manufactured products and currently recognizes 67 Islamic bodies in 41 countries around the world as certification bodies for products imported into Malaysia. Therefore, a more practical way is required to assist Muslims in recognizing various forms of Halal logos on food packaging. A neural network (NN) approach is proposed to recognize authentic and recognized Halal logo on imported products. A dataset of available and recognized Halal logo images worldwide will be created for this purpose. The dataset will be used to train and test the performance of the learning algorithm to recognize logo of recognized foreign bodies by JAKIM. The approach is expected to complement current facilities for verification using Short Messaging Services (SMS) and web portal. The approach is assumed to be more efficient and accurate for Halal logo verification which eventually could win the trust of Halal product consumers and support the Halal industry in Malaysia.</span>

Connecting the dots between brand logo and brand image

PurposeAlthough the prominence of brand logo for companies is widely acknowledged, a close examination of the literature reveals lack of empirical research pertaining to effect of brand logo on consumer perception toward brand. Therefore, the purpose of this paper is to fill the gap in marketing studies concerning the effect of a logo on consumer evaluations. The research addresses two questions: first, how brand logo favorability helps to increase brand image; and second, how brand logo form consumer perceptions toward brand through brand personality dimensions and brand familiarity?Design/methodology/approachThe data were collected from 816 respondents using mall-intercept technique. Structural equation modeling via AMOS was conducted to test the proposed model to gain insight into the various relevant influences and relationships.FindingsThe findings revealed the importance of the company’s brand logo in enhancing the brand image. The results further highlighted that brand personality dimensions and brand familiarity mediate the relationship between brand logo and brand image.Practical implicationsThe study offers managers a new perspective for building strong brand identity with the help of logos along with the brand personality dimensions and brand familiarity to enhance brand image.Originality/valueThis study provides novel insights on the impact of brand logo on brand image. This is the first study to determine the mediating role of brand personality in the relationship between brand logo and brand image. It thereby adds to the literature of visual identity by developing the sphere of influence of brand logo and its effects toward brands.

Preventing phishing attacks using text and image watermarking

SummaryPhishing is a form of identity theft used to illegally gather personal and financial information by providing fake web pages designed to mimic the website of a legitimate business. In this paper, we propose a novel anti‐phishing approach based on using imperceptible watermarking and monitoring the HTTP requests. In the server‐side, a URL dependent watermark is generated and embedded in the elements of the web page, which are logo image and HTML/CSS files. In the client side, to authenticate the website and check its safety, the watermark is regenerated and compared with the extracted watermarks from the elements of the webpage. Security analysis of the proposed method demonstrates its robustness against some common attacks. The proposed method performs fully automatically and without user interaction. It can be simply implemented and used for all kinds of websites, either secured or not secured with SSL protocol.

Hybrid pyramid-DWT-SVD dual data hiding technique for videos ownership protection

This paper proposes a hybrid pyramid Discrete-Wavelet-Transform (DWT) Singular-Value-Decomposition (SVD) data hiding scheme for video authentication and ownership protection. The data being hidden will be in the shape of a main color logo image watermark and another secondary Black and White (BW then, a Discrete Wavelet Transform (DWT) process is implemented using orthonormal filter banks on these error images, and the Bit-Slices watermarks are inserted in one or more of the resulting subbands in a way that is fully controlled by the owner; then, the watermarked video is reconstructed. SVD will be performed on the color watermark Bit-Slices. A secondary BW meanwhile, the mean Peak Signal-to-Noise Ratio (PSNR) is 57 dB.

How hand gestures influence the enjoyment in gamified mobile marketing

Abstract With the rapid development of mobile technologies, gamification has been deployed pervasively on mobile platforms as an effective marketing tool. In this study, we focus on how mobile gesture technology influences consumer enjoyment of mobile marketing games by examining two types of hand gestures: surface and motion gestures. Considering the characteristics of hand gestures and embodied cognition theories, we propose that hand gestures influence the enjoyment of mobile marketing games with two other game elements: object visual presentation and reward setting. Specifically, the interaction between surface gesture (vs. motion gesture) and the object visual presentation of a real product picture (vs. symbolic brand logo image) leads to greater enjoyment of marketing games through mental simulation. Similarly, the interaction between motion gesture (vs. surface gesture) and reward setting of uncertainty (vs. certainty) leads to greater enjoyment of marketing games through perceived control. Three online experiments are conducted to support the proposed hypotheses. Results provide implications for marketing practitioners that hand gesture design should be aligned with congruent object visual presentation and reward setting in order to enhance consumer enjoyment of mobile marketing games. This study contributes to the extant literature of gamification, as well as human-computer interaction.

Ternary radial harmonic Fourier moments based robust stereo image zero-watermarking algorithm

With the development and popularization of computer network technology, the copyright protection of stereo images is a serious problem to be solved. Based on ternary number theory and radial harmonic Fourier moments (RHFM), ternary radial harmonic Fourier moments (TRHFM) is proposed to deal with stereo images in a holistic manner, and based on this moment, this paper proposes a robust stereo image zero-watermarking algorithm. We first compute the TRHFM of the original stereo image, and we randomly select TRHFMs using logistic mapping; then, we obtain a binary feature image using the magnitudes of the selected TRHFMs, and finally, we apply a bitwise exclusive-or operation on permuted logo image and binary feature image to obtain the zero-watermark image. Experimental results indicate that the proposed stereo image zero-watermarking algorithm is strongly robust to various asymmetric and symmetric attacks and has superiority compared with other zero-watermarking algorithms.

Vehicle logo recognition based on overlapping enhanced patterns of oriented edge magnitudes

Vehicle logo recognition (VLR) has attracted wide attention from the community of intelligent transportation systems (ITS) due to its important role. Although many methods have been proposed for VLR, it remains a challenging problem. In this paper, we present a novel method for VLR. Our method includes (1) observation of the local anisotropy of vehicle logo images; (2) adoption of the idea of patterns of oriented edge magnitudes (POEM) and an advanced version of POEM for vehicle logo feature description called overlapping enhanced POEM (OE-POEM); (3) implementation of whitened principal component analysis (WPCA) for feature dimension reduction followed by collaborative-representation-based classification (CRC) as a classifier to perform VLR. We also construct a new vehicle logo dataset (HFUT-VL), which is larger and more comprehensive than the existing vehicle logo datasets. Finally, we conduct experiments on HFUT-VL, and the results indicate state-of-the-art VLR performance.