Уровень 0 · материалов: 3
В кластер входят документы, описывающие применение алгоритмов распознавания образов или биометрического сканирования для идентификации кошек, и не входят документы об общих методах борьбы со спамом без привязки к анализу изображений животных.
Общие признаки: использование технологий компьютерного зрения, идентификация кошек по изображениям, технические эксперименты
Группа выше: Распознавание лиц и объектов
Смысл: The text is a lighthearted technical experiment using image recognition (Caffe) to count how many users on the Habr platform use cat images as their avatars, illustrating the practical application and limitations of deep learning.
An author used the Caffe deep learning framework to analyze over 500,000 Habr profiles, ultimately identifying 745 users with cat avatars.
Смысл: The text is a brief technical post where a blogger shares their implementation of an image-based CAPTCHA (distinguishing cats from dogs) to stop spam, while seeking community feedback to improve it.
A developer implements a simple image-recognition CAPTCHA to stop blog spam and asks the community for feedback to make it more secure.
Смысл: The text is a lighthearted April Fools' experiment exploring whether consumer-grade biometric scanners can identify individual cats via their noses, concluding that while technically possible, it is practically unfeasible due to cat behavior.
An engineer attempts to build a biometric door lock for his cat using a fingerprint scanner, finding that while noses can be scanned, cats are too uncooperative for it to work.