Trademark Recognition Research Articles

Новые вызовы в применении технологий искусственного интеллекта: нарушение прав на товарные знаки на платформах электронной коммерции

The article explores new challenges associated with the use of artificial intelligence technologies in the context of trademark infringement on e-commerce platforms. In the modern digital world, where ecommerce is growing rapidly, issues of intellectual property protection, including trademark rights, are becoming especially relevant. Various scenarios in which trademark violations on e-commerce platforms are possible are analyzed, and how artificial intelligence technologies can be used to identify and prevent such violations are considered. The possibilities of trademark recognition and classification using machine learning algorithms and neural networks are being investigated. The problems associated with false positive and false negative results when using artificial intelligence in this context are discussed. Recommendations and principles are proposed for the development of effective trademark protection systems on ecommerce platforms that will take these issues into account and ensure fairness and compliance with legislation. The importance of the development and application of artificial intelligence technologies to combat trademark violations on e-commerce platforms is emphasized, the importance of developing these technologies taking into account special challenges and requirements in this area, in order to ensure the protection of intellectual property rights and support fair competition in the online environment is emphasized.

Intelligent trademark recognition and similarity analysis using a two-stage transfer learning approach

The ubiquitous and fast-growing e-marketplaces are causing serious concerns of unauthorized trademark usages, especially their pictorial images. There is a significant need for e-commerce service intermediaries and trading sites to actively check whether product images posted online are not infringing upon others’ intellectual property rights (IPRs). To ensure registered trademark (TM) or copyrighted logo-image (logo) protection, this research develops an intelligent system that can detect, locate, and crop (cut) logos posted online and check whether they are substantially or deceptively similar to registered TM logos. This research integrates two deep learning models to achieve the research goal. The first is the logo detection and localization model for cropping trademark like images from complex online merchandise photos, which could have images of many product views and even worn and displayed on a human model. Only the cropped logo image is needed for comparison to a database of registered trademarks. The second model performs TM similarity analysis using the cropped logos compared to the published TM images. Yolo v4 is adopted as the general logo locator for intelligent logo image cropping. The triplet convolutional neural network model is used to fine-tuned for intelligent trademark similarity analysis. The models are trained with an image dataset, combining image samples from LogoDet-3k and images found through web search. The system performance is quantitatively tested using the FlickrLogos-32 dataset and consumer product images extracted from an e-commerce platform. The test results of the system achieve high precision (0.91) for trademark class matching.

Application Research of Trademark Recognition Technology based on SIFT Feature Recognition Algorithm in Advertising Design

Now-a-days, due to the sharp increase in the number of advertising designs, creative duplication is easy to occur in advertising design. If this situation is not discovered in time, it may cause legal disputes and cause damage to the reputation and property of the enterprise. In view of the above situation, this paper proposes a trademark recognition technology based on SIFT feature recognition algorithm to avoid duplication of advertisement design and cause copyright disputes. Aiming at the defect that the dimension of the image feature vector extracted by SIFT algorithm is too high, the principal component analysis method is used to reduce its dimension. For the problem of unsatisfactory image recognition rate accuracy of SIFT algorithm, a support vector machine is used to classify the extracted feature vector, so as to improve the image recognition rate. Based on the above content, build a trademark recognition model. The research results show that the recognition accuracy of the model reaches 98.82%, 0.66% and 0.58% higher than that of Model 1 and Model 2; AUC value of model 3 is 0.962, 0.039 higher than model 2 and 0.107 higher than model 1. The above results show that the proposed trademark recognition model can better identify similar advertising designs, thereby avoiding design duplication and legal disputes.

Authentication of Transylvanian Spirits Based on Isotope and Elemental Signatures in Conjunction with Statistical Methods.

The potential association between stable isotope ratios of light elements and mineral content, in conjunction with unsupervised and supervised statistical methods, for differentiation of spirits, with respect to some previously defined criteria, is reviewed in this work. Thus, based on linear discriminant analysis (LDA), it was possible to differentiate the geographical origin of distillates in a percentage of 96.2% for the initial validation, and the cross-validation step of the method returned 84.6% of correctly classified samples. An excellent separation was also obtained for the differentiation of spirits producers, 100% in initial classification, and 95.7% in cross-validation, respectively. For the varietal recognition, the best differentiation was achieved for apricot and pear distillates, a 100% discrimination being obtained in both classifications (initial and cross-validation). Good classification percentages were also obtained for plum and apple distillates, where models with 88.2% and 82.4% in initial and cross-validation, respectively, were achieved for plum differentiation. A similar value in the cross-validation procedure was reached for the apple spirits. The lowest classification percent was obtained for quince distillates (76.5% in initial classification followed by 70.4% in cross-validation). Our results have high practical importance, especially for trademark recognition, taking into account that fruit distillates are high-value commodities; therefore, the temptation of “fraud”, i.e., by passing regular distillates as branded ones, could occur.

Novel Relevance Feedback Approach for Color Trademark Recognition Using Optimization and Learning Strategy

Abstract The trademark registration process, apparent in all organizations nowadays, deals with recognition and retrieval of similar trademark images from trademark databases. Trademark retrieval is an imperative application area of content-based image retrieval. The main challenges in designing and developing this application area are reducing the semantic gap, obtaining higher accuracy, reducing computation complexity, and subsequently the execution time. The proposed work focuses on these challenges. This paper proposes the relevance feedback system embedded with optimization and unsupervised learning technique as the preprocessing stage, for trademark recognition. The search space is reduced by using particle swam optimization, for optimization of database feature set, which is further followed by clustering using self-organizing map. The relevance feedback technique is implemented over this preprocessed feature set. Experimentation is done using the FlickrLogos-32 PLUS dataset. To introduce variations between the training and query images, transformations are applied to each of the query image, viz. rotation, scaling, and translation of the image. The same query image is tested for various combinations of transformations. The proposed technique is invariant to various transformations, with significant performance as depicted in the results.

Reclaiming the Primary Significance Test: Dictionaries, Corpus Linguistics, and Trademark Genericide

Companies spend billions to promote their brand. But with increased trademark recognition comes the possibility of losing exclusive rights to use that trademark, a process called genericide. Courts have often turned to linguistic evidence, such as dictionaries and media usage, to determine whether a trademark has become generic. These courts merely suggest that linguistic tools reflect a trademark’s meaning. Yet these tools are not the objective indicators that courts have assumed. This Paper discusses why using dictionaries and media usage to prove genericide is a mistake and then turns to evaluating another tool, corpus linguistics. Corpus linguistics, unlike other linguistic tools, may prove beneficial for companies seeking to protect their trademarks. Ultimately, however, linguistic tools — including dictionaries, media usage, and corpus linguistics — cannot prove genericism because linguistic data may, at best, prove a term’s majority usage. But the Lanham Act requires a showing of primary significance. The Paper contends that courts should maintain majority usage and primary significance as distinct concepts and, in this way, should reclaim the primary significance test.

Exemplar-based logo and trademark recognition

Automatic logo recognition facilitates the process of searching and analyzing digital image-based resources. This is due to the simplicity and informativity of logos which can be used as a visual or semantic clue for applications such as content-based image retrieval. For automatic shape recognition, model-based approaches have gained a lot of attention due to their theoretical background and promising results. However, one of the major challenges of these approaches for logo recognition is collecting samples of logos with possible perspective tilts and non-rigid deformations that most likely happen to natural images. Therefore, a robust model cannot be trained due to the lack of sufficient samples. In this paper, we aim to overcome, or at least alleviate, this challenge by introducing an exemplar-based approach. The proposed method synthesizes new images of each logo of interest, so-called exemplars, by simulating two camera axes. This is to produce almost all possible views of a logo that may appear in a given image and guarantee that most of the distortions in the shape of a logo caused by a variation of the camera optical axis direction are covered. Subsequently, an exemplar-based model is derived based on all the exemplars using linear SVMs. Finally, for the detection and recognition, we associate observed logos with the exemplars using our exemplar-based model. Experiments are carried out on two publicly available logo datasets: FlickrLogos-27 and FlickrLogos-32. The results show the superior detection and recognition capabilities of the proposed method over the competitors.

How are Unregistered Trademarks Protected in China?

As in most civil law countries, registration is the only way to acquire trademark rights in China, and the protection of unregistered marks remains a marginal issue. However, companies always have considerable interests in the protection of their unregistered commercial signs. Problems with trademark piracy and passing off regarding unregistered trademarks in China have drawn attention throughout the world. Besides the absence of spontaneous market order, defects in the current trademark and competition law systems are the main reasons for this problem. The bona fide principle, as incorporated in the PRC Trademark Law, is applicable only to piracy of those trademarks that have been used in China and are known to the relevant public. The action of passing off in the PRC Anti-Unfair Competition Law does not cover unregistered trademarks as it stands, and they can only benefit from this protection by being disguised as a product name, packing or decoration. The protection of well-known trademarks is nearly unavailable because the fame requirement is too high. The last revision of the PRC Trademark Law improved the current regime by enlarging the scope of application of the bona fide principle in trademark law. However, we still need to give a clear status to unregistered trademarks in anti-unfair competition law, and Chinese courts should lower the fame requirement in the recognition of well-known trademarks.

Trademark Recognition Based on Hu Modified Invariant Moments

Trademark represents the quality of the goods and the manufacturer's reputation, which is an important symbol of goods; automatic trademark recognition has a wide range of application prospects in many fields such as e-commerce, advertising and logistics transportation. In target recognition algorithm, the most critical is to extract the target feature. Because of the moment feature has better invariant characteristics; it is widely used in image feature extraction. In this paper we use Hu modified moments and the minimum distance classifier for trademark recognition, the experimental results show that the modified moments have higher recognition rate than the original Hu moments, this method can improve the recognition rate of the target image, it is conducive to trademark recognition.

A Real-time Billboard Trademark Recognition Algorithm in Sports Video

摘要: 近期, 体育视频分析中的广告牌商标的探测和识别方法已经广泛应用于许多其他领域, 比如商业电视. 基于此, 提出了一种能在不同体育视频(如足球、篮球和F1赛车等)中进行广告牌商标实时识别的算法, 该算法主要包括两个步骤, 首先, 利用基于模糊决策树的方法进行广告牌图像帧的探测; 其次, 利用颜色特征和局部SIFT (Scale-invariant feature transform)特征来描述不同商标的外观, 并最终通过基于潜在语义分析(Latent semantic analysis, LSA)的SIFT词汇匹配来识别所给定的商标模板. 初步的实验表明了本文算法的有效性, 并且该算法能在实时情况下运行. 关键词: 商标识别 / 模糊决策树 / SIFT词汇 / 潜在语义分析

Creating own Brand Equity Versus Customer Loyalty in a Commercial Enterprise

A dynamic global market aims at creating brands (including own brands) on geographically and structurally new markets, which considerably influences the perception and improvement of their image. The degree of trademark recognition within a given line of business is the key factor in brand management planning. Owing to new technologies and reduction in manufacturing costs, producers have access to additional productive capacity, which is used by external entities (including commercial enterprises) to create their own brands with their own image and identity, which often differs from the concept and strategy of the manufacturer brand.

The recognition and protection of well-known trade marks in Indonesia

Legal context The recognition and protection of well-known marks in Indonesia has improved over the last few years for a variety of reasons. Key points First, the Asian Crisis resulted in the creation of a Commercial Court, which is a clear improvement over the previously responsible District Courts. Secondly, the increasingly frequent publication of court decisions has improved transparency and consistency of those decisions. Well-known marks are now clearly protected against use for similar goods/services. Protection is extended to dissimilar goods/services by applying Article 16(3) TRIPS directly or by arguing that registration occurred in bad faith. Nevertheless, decisions thus far concern almost exclusively revocation and invalidity of registrations. Civil remedies such as damages and interim injunctions are hardly used, because the outdated civil procedural law has not familiarised judges with such legal instruments. Clearing the register of infringing registrations is another matter of concern. Cancellation for non-use for three consecutive years can be difficult, because the plaintiff is required to provide evidence of the last use in the production of the goods/services rather than in the course of trade more generally. Practical significance While it has become much easier to protect well-known marks in Indonesia, much work remains to be done regarding the procedural framework of civil infringement proceedings and regarding the clearing of the register.

The Illegitimate Use of Trade Symbols as an Infringement of the Unfair Competition Law in Colombia

The reputation of a firm in the market is identified by the customers through symbols that convey information about the quality of the products, such as trade names, trade marks and other trade symbols. Companies invest important resources to build a good reputation regarding its products and services; hence, the strength, significance and recognition of trade marks, trade names and corporate names are important assets for a firm in any market. The objective of this document is to explain the protection of trade symbols in the Colombian legislation and case-law. The document focuses on the protection of trade names, corporate names and the names of commercial premises and the prohibition - by the Unfair Competition Law - of two specific unfair competition types: likelihood of confusion and profit from another's reputation.

Trademark shape recognition using closed contours

In this paper, a new trademark recognition method using closed contours is proposed. The developed method can be used to retrieve similar trademarks, even those with irregular shapes, so as to simplify and speed up the manual trademark examination process. Trademarks are first decomposed into complete sets of elementary closed contours, each of which is then encoded as an angle-code string according to chain-code information. A two-step string-matching algorithm can then be employed to compute similarities between closed contours. Finally, the maximum and average terms of the similarities between contours are integrated into the whole trademark similarity measure. After that, the similarity measure can be the criterion for retrieving similar trademarks. More specifically, since the proposed method to encode closed contours is insensitive to translation, rotation and size variations, our method has the ability to detect infringement cases in which trademarks are changed in translation, rotation, or size. Moreover, our method can cope with the special resemblances such as inclusion, mirror and partial resemblances. Experimental results are presented to show the feasibility and practicability of the proposed approach.