The development of artificial intelligence (AI) has revolutionized the capacity to address complex machine intelligence problems. Machine learning (ML) is currently a big challenge because it directly trains machines without much human intervention, the days when data had to be entered into the system manually are gone; today’s systems learn automatically. Supervised and unsupervised ML techniques such as feature extraction, object detection, classification, and pattern recognition are used for various purposes. In the field of computer vision (CV), ML plays a vital role in extracting meaningful details came based on visual representations, which in turn have applications ranging from surveillance systems to robotics or even suspect identification using facial recognition based on images captured from various sources. Healthcare is a leading CV research area with a focus on medical imaging (MI). Thus, MI can be considered a secondary contribution of artificial intelligence to healthcare, in addition to other fields such as OCR or robotics. MI is a promising technology that plays a huge role in image enhancement. In particular, the process involves transforming the original image into gray scale and finding optimal threshold values for segmentation. Additionally, the presence of previous work restrictions and potential obstacles to future work are acknowledged. Moreover, we identify and examine numerous significant obstacles that need to be resolved in order to efficiently utilize machine learning in the fields of computer vision and image processing. This review paper examines various supervised and unsupervised machine learning (ML) algorithms, provides a broad overview of image processing and its implications, discusses neural network enabled models, constraints, tools, and computer vision (CV) applications, and finally, emphasizes important areas of open research in ML for CV.