Disordered eating attitudes are rapidly increasing, especially among young women in their twenties. These disordered behaviours result from the interaction of several factors, including beauty ideals. A significant factor is social media, by which the unrealistic beauty ideals are popularized and may lead to these behaviours. The objectives of this study were, first, to determine the relationship between disordered eating behaviours among female university students and sociocultural factors, such as the use of social network sites, beauty ideals, body satisfaction, body image and the body image desired to achieve and, second, to determine whether there is a sensitive relationship between disordered eating attitudes, addiction to social networks, and testosterone levels as a biological factor. The data (N = 168) was obtained using validated surveys (EAT-26, BSQ, CIPE-a, SNSA) and indirect measures of prenatal testosterone. The data was analysed using chi-square, Student’s t-test, correlation tests and logistic regression tests. The results showed that disordered eating attitudes were linked to self-esteem (p < 0.001), body image (p < 0.001), body desired to achieve (p < 0.001), the use of social media (p < 0.001) and prenatal testosterone (p < 0.01). The findings presented in this study suggest a relationship between body image, body concerns, body dissatisfaction, and disordered eating attitudes among college women.

Three-dimensional photo-reconstruction (PR) techniques have been successfully used to produce high-resolution surface models for different applications and over different spatial scales. However, innovative approaches are required to overcome some limitations that this technique may present for field image acquisition in challenging scene geometries. Here, we evaluate SF3M, a new graphical user interface for implementing a complete PR workflow based on freely available software (including external calls to VisualSFM and CloudCompare), in combination with a low-cost survey design for the reconstruction of a several-hundred-metres-long gully network. SF3M provided a semi-automated workflow for 3-D reconstruction requiring ~ 49 h (of which only 17 % required operator assistance) for obtaining a final gully network model of > 17 million points over a gully plan area of 4230 m2. We show that a walking itinerary along the gully perimeter using two lightweight automatic cameras (1 s time-lapse mode) and a 6 m long pole is an efficient method for 3-D monitoring of gullies, at a low cost (~ EUR 1000 budget for the field equipment) and the time requirements (~ 90 min for image collection). A mean error of 6.9 cm at the ground control points was found, mainly due to model deformations derived from the linear geometry of the gully and residual errors in camera calibration. The straightforward image collection and processing approach can be of great benefit for non-expert users working on gully erosion assessment.