In industries, the detection of anomalies such as scratches, dents, and discolorations is crucial to ensure product quality and safety. However, conventional methods rely on heavy computational processing and image enhancement and may not truly reflect subtle defects, particularly in low-light settings. Now, researchers have designed a robust model with noise suppression and illumination-adaptive features that enhance the accuracy and consistency of anomaly detection across diverse surfaces and textures in poorly lit industrial environments.

A team of biologists and robotic engineers have developed a virtual reality system for fish to decipher how they school. They uncovered the natural ‘control law’ that is used by zebrafish to coordinate behaviour with others, a behavioural algorithm that has been tuned over millennia to facilitate effective collective motion. Since such collective behaviour could be highly advantageous for technological solutions, such as in the control of autonomous vehicles, the scientists then tested its performance with groups of robotic cars, drones and watercraft. They found that the rules of interaction that have evolved in fish are highly effective for robotic control, demonstrating their potential for the control of fleets of robots in the future. The research, published in Science Robotics, was led by scientists at the Cluster of Excellence Collective Behaviour, University of Konstanz, and the Max Planck Institute of Animal Behavior (MPI-AB), Germany, in collaboration with researchers at Eötvös University in Hungary and the Massachusetts Institute of Technology (MIT) in the United States.

Researchers led by the University of Cincinnati’s Anna Kruyer and the University of Houston’s Demetrio Labate have published research in the journal Science Advances applying object recognition technology to track changes in brain cell structure and provide new insights into how the brain responds to heroin use, withdrawal and relapse.