Background: Maize cobs are an important component of crop yield that exhibit a high diversity in size, shape and color in native landraces and modern varieties. Various phenotyping approaches were developed to measure maize cob parameters in a high throughput fashion. More recently, deep learning methods like convolutional neural networks (CNN) became available and were shown to be highly useful for high-throughput plant phenotyping. We aimed at comparing classical image segmentation with deep learning methods for maize cob image segmentation and phenotyping using a large image dataset of native maize landrace diversity from Peru. Results: Comparison of three image analysis methods showed that a Mask R-CNN trained on a diverse set of maize cob images was highly superior to classical image analysis due to its robustness to image quality and object segmentation accuracy (r=0.99). The Mask R-CNN was integrated into a high-throughput pipeline that segments both maize cobs and a ruler in images and then performs automated quantitative analysis of eight phenotypic traits, including diameter, length, ellipticity, asymmetry, aspect ratio and average RGB values for cob color. Statistical analysis identified key training parameters for efficient iterative model updating. Only a small number of 10-20 images are required to update the initial Mask-R-CNN model to process new types of cob images. We demonstrate an application of the pipeline by analyzing phenotypic variation in 19,867 maize cobs contained in 3,449 images of 2,484 accessions from the maize genebank of Peru to identify phenotypically homogeneous and heterogeneous genebank accessions using multivariate clustering. Conclusions: We show that a single Mask R-CNN model and associated analysis pipeline are widely applicable tools for maize cob phenotyping in contexts like genebank phenomics or plant breeding.