Abstract: In a method for recovering lithium hydroxide from a lithium secondary battery, cathode powder is prepared from a cathode of the lithium secondary battery. A cathode active material mixture is prepared by mixing the cathode powder with a calcium compound. The cathode active material mixture is reduced to form a preliminary precursor mixture. A lithium precursor is recovered from the preliminary precursor mixture. Therefore, a lithium precursor can be obtained with high purity without a complicated leaching process or an additional process, which result from a wet-based acid solution process.

Abstract: A lens module assembly optimization method includes: in preparing a lens module including assembled N lenses respectively formed in cavities: receiving characteristic information of at least N lenses respectively formed in N cavity groups each including a respective plurality of cavities; and processing information for selecting N cavities from the N cavity groups, based on the characteristic information. A past cavity selection result, a fitness function configured based on data of the assembled N lenses or data of the prepared lens module according to the past cavity selection result, and a genetic algorithm are received or stored.

Abstract: A deep learning-based MLCC stacked alignment inspection system includes an integrated defect detection unit configured to detect core areas requiring inspection of image data in which a stacked structure is photographed from a semiconductor MLCC chip by using at least one deep learning-based core area detection model, perform segmentation in the detected core areas, determine whether a defect exists according to a standard margin percentage range, and enable defect detection by generating normal and/or defective data based on the determination result, a result analysis unit configured to perform visualization for respective results of the core area detection, segmentation, and defect detection of the integrated defect detection unit, and provide stepwise analysis data for the visualized respective results so as to determine whether to modify corresponding data, and a data storage configured to store the normal and/or defective data, and stepwise analysis data.