NETWORK CLASSIFIER FOR IMAGE RECOGNITION

Image recognition is a fundamental task in computer vision that entails mapping visual data to discrete labels. Inthis report, we survey different network classifiers for image recognition, evaluate several case studies, anddiscuss ethical considerations surrounding their deployment. We focus on convolutional neural networks(CNNs), traditional machine-learning classifiers such as support vector machines (SVMs) and random forests,and illustrate their strengths and weaknesses through empirical studies. We also highlight ethical issues—including bias, data privacy, transparency, autonomy,and accountability—that must be addressed whendeploying these models.