# Evaluating Contextual Bias in CNN Image Classification: Evidence from Agricultural Benchmark Datasets

arXiv: https://arxiv.org/abs/2609.14654 · HTML: https://mlsanity.com/paper/2609.14654/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 19 of its 42 references that are in the map.
- First posted on arXiv 13 Sep 2026.

## Builds on (most cited first, 19 of 19)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Very Deep Convolutional Networks for Large-Scale Image Recognition](https://mlsanity.com/paper/1409.1556/) (2014)
- [Densely Connected Convolutional Networks](https://mlsanity.com/paper/1608.06993/) (2016)
- [Rethinking the Inception Architecture for Computer Vision](https://mlsanity.com/paper/1512.00567/) (2015)
- ["Why Should I Trust You?": Explaining the Predictions of Any Classifier](https://mlsanity.com/paper/1602.04938/) (2016)
- [Identity Mappings in Deep Residual Networks](https://mlsanity.com/paper/1603.05027/) (2016)
- [ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustness](https://mlsanity.com/paper/1811.12231/) (2018)
- [Recognition in Terra Incognita](https://mlsanity.com/paper/1807.04975/) (2018)
- [Confounding variables can degrade generalization performance of radiological deep learning models](https://mlsanity.com/paper/1807.00431/) (2018)
- [Using Deep Learning for Image-Based Plant Disease Detection](https://mlsanity.com/paper/1604.03169/) (2016)
- [DeepWeeds: A Multiclass Weed Species Image Dataset for Deep Learning](https://mlsanity.com/paper/1810.05726/) (2018)
- [PlantDoc: A Dataset for Visual Plant Disease Detection](https://mlsanity.com/paper/1911.10317/) (2019)
- [Not Using the Car to See the Sidewalk: Quantifying and Controlling the Effects of Context in Classification and Segmentation](https://mlsanity.com/paper/1812.06707/) (2018)
- [A Survey of Deep Learning Techniques for Weed Detection from Images](https://mlsanity.com/paper/2103.01415/) (2021)
- [A Survey on Understanding, Visualizations, and Explanation of Deep Neural Networks](https://mlsanity.com/paper/2102.01792/) (2021)
- [Using Transfer Learning for Image-Based Cassava Disease Detection](https://mlsanity.com/paper/1707.03717/) (2017)
- [Systematic biases when using deep neural networks for annotating large catalogs of astronomical images](https://mlsanity.com/paper/2201.03131/) (2022)
- [Analyzing the Impact of Shape & Context on the Face Recognition Performance of Deep Networks](https://mlsanity.com/paper/2208.02991/) (2022)
- [Unmasking Biases and Reliability Concerns in Convolutional Neural Networks Analysis of Cancer Pathology Images](https://mlsanity.com/paper/2603.12445/) (2026)

## Abstract (arXiv)

Convolutional neural networks (CNNs) are typically evaluated using held-out classification accuracy, an approach that presupposes predictions are based primarily on the intended object of interest rather than incidental surrounding context. We test this assumption in CNN-based agricultural image classification by comparing model performance on original images with performance on background-dominated patches extracted from the same images across eight publicly available agricultural benchmark datasets and four widely used CNN architectures. Background-dominated patches were classified above dataset-specific random chance for six of the eight datasets, and substantially above chance for four of them, indicating that contextual information contributes to model predictions for the majority of datasets evaluated. For these four datasets, we further evaluated whether this behavior reflected genuine class-discriminative information or was primarily attributable to class imbalance using macro-averaged precision, recall, and F1 together with class-balanced test subsets. The results show that contextual reliance does not admit a single explanation: class imbalance accounts for a substantial portion of the observed signal for some datasets and architectures, whereas above-chance contextual classification persists after balancing for others. Together with previous evidence from curated object recognition and cancer pathology imaging, these findings support the growing view that contextual bias is a recurring characteristic of CNN-based image classification rather than a phenomenon confined to a single application domain. More broadly, this work provides a systematic framework for quantifying contextual bias across heterogeneous image datasets by combining dataset-specific random-chance baselines, contextual bias categorization, macro-averaged evaluation, and class-balanced robustness analysis.