Convolutional Neural Networks (CNNs) operate within a wide variety of hyperparameters, the optimization of which can greatly improve the performance of CNNs when performing the task at hand. However, these hyperparameters can be very difficult to optimize, either manually or by brute force. Neural architecture search or NAS methods have been developed to address this problem and are used to find the best architectures for the deep learning paradigm. In this article, a CNN has been evolved with a well-known nature-inspired metaheuristic paddy field algorithm (PFA). It can be seen that PFA can evolve the neural architecture using the Google Landmarks Dataset V2, which is one of the toughest datasets available in the literature. The CNN’s performance, when evaluated based on the accuracy benchmark, increases from an accuracy of 0.53 to 0.76, which is an improvement of more than 40%. The evolved architecture also shows some major improvements in hyperparameters that are normally considered to be the best suited for the task.
AI image processing has emerged as a transformative tool in the realm of maternal healthcare, particularly in the early detection of pregnancy complications. By harnessing the power of artificial intelligence, healthcare providers can now leverage advanced algorithms to analyze medical images such as ultrasound scans, MRI images, and fetal monitoring data with unprecedented accuracy and efficiency. These AI-based systems excel at detecting subtle abnormalities and anomalies that may indicate potential risks to maternal and fetal health, including markers of fetal growth restriction, placental abnormalities, and congenital anomalies. By facilitating earlier detection and intervention, AI image processing empowers healthcare providers to proactively manage pregnancy complications, thereby improving outcomes for both mother and baby. These predictive models enable healthcare providers to assess the risk of complications early in pregnancy and tailor interventions accordingly.
Recognizing landmarks in images with machine learning is an excellent topic for research today. Landmark recognition is an important field in computer vision. In this field, we train the machine learning models to identify and recognize the closed distinctly distinguishable objects in a digital image. In general, if we consider a digital image to be a set of coordinates of different pixels, a landmark is said to be enclosed in that closed polygon formed by the pixels that may be considered as a distinct and distinguishable thing in one or the other sense. Landmark recognition is an important subject area of image classification since it is considered as one of the first steps towards reaching complete computer vision. The extremely broad definition of a landmark makes it eligible to be considered as one of the leading problems in image classification tasks. Since the task is considered to be a very broad one, the solutions to the task hold no easy procedures. This chapter explores landmark recognition using ensemble-based machine learning models.