相关实验视频

Updated: Mar 9, 2026

Author Spotlight: Assessment of Visual Acuity in Central Vision Loss Through Motion-Based Peripheral Vision Testing

Author Spotlight: Assessment of Visual Acuity in Central Vision Loss Through Motion-Based Peripheral Vision Testing

Published on: February 23, 2024

RAC:通过基于CLIP的模两可的减少来识别近距离的果实

Dat Tran-Anh¹, Thang Vu Ba², Ngan Hoang Dao³

¹Graduate University of Science and Technology, Vietnam Academy of Science and Technology, Hanoi, Vietnam; Faculty of Information Technology, Thuyloi University, Hanoi, Vietnam.

Neural networks : the official journal of the International Neural Network Society

|March 7, 2026

概括

这项研究引入了一个新的框架,降低Clip中的模两可,用于假冒检测 (RAC),以改善假冒农产品的识别. RAC有效地减少了视觉模两可,在短暂的学习场景中显著提高了检测准确性.

科学领域:

计算机视觉计算机视觉
机器学习机器学习
农业科学农业科学

背景情况:

假冒农产品带来了重大分类挑战,特别是像CLIP这样的模型在少数射击学习条件下.
真正和假冒物品之间的微妙视觉差异会造成模糊性,阻碍准确的检测.

研究的目的:

引入一个新的框架,减少伪造检测CLIP (RAC) 中的模糊性,以解决伪造农产品检测中的模糊性.
增强歧视力,减少在短暂学习场景中的类间混.

主要方法:

该RAC框架使用多层感知器Plus (MLP-P) 模块来合成来自CLIP图像编码器的多层视觉表示.
使用一个类间模糊性减少 (IRA) 模块来抑制导致混的视觉模式.
一个网络融合 (NF) 阶段动态集成输出,以便进行可靠的分类.

主要成果:

在四个基准测试中,RAC在16次学习中实现了76.95%的平均准确率,在四个基准测试中超过了最先进的2.50%.
该框架在TFS-Fruit数据集上达到80.0%的准确性,证明了其在关键农业应用中的有效性.

结论:

该RAC框架显著提高了对假冒农产品检测的准确性,特别是在短暂的学习场景中.
RAC的活跃提炼过程有效地减轻了视觉模两可,为现实世界农业挑战提供了强大的解决方案.

关键词:

农业假货检测农业假货检测有几次射击学习学习.果实识别功能果实识别功能图像处理图像处理

更多相关视频

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

Fruit Volatile Analysis Using an Electronic Nose

Fruit Volatile Analysis Using an Electronic Nose

Published on: March 30, 2012

相关概念视频

Fruit Development, Structure, and Function

Fruit Development, Structure, and Function

Fruits form from a mature flower ovary. As seeds develop from the ovules contained within, the ovary wall undergoes a series of complex changes to form fruit. In some fruits, such as soybeans, the ovary wall dries; in other fruits, such as grapes, it remains fleshy. In some cases, organs other than the ovary contribute to fruit formation; such fruits are called accessory fruits.