Search

BACKGROUND Gypsum-salt formation affects the oxygen fugacity of ore-forming fluid and changes the fluid composition, and then affects the trace element composition, which plays an important indicator role in the formation of iron ore deposits. Therefore, changes in the elemental composition of magnetite can be used to classify the types of deposits and determine their genesis. Gypsum-salt formation is widely developed in "Pamir-type" iron deposits in Xinjiang and Ningwu iron deposits in the middle and lower part of the Yangtze River. However, the controlling mechanism of gypsum-salt formation is still unclear. OBJECTIVES In order to investigate the metallogenic process and formation environment of the two types of magnetite, and to discuss the role of gypsum-salt formation in the formation of magnetite deposits. METHODS In situ LA-ICP-MS were employed to determine trace elements in magnetite. RESULTS: LA-ICP-MS results showed that magnetite in the Ningwu area were mainly rich in Ti and V, indicating that it was closely related to magmatism, while the contents of Nb, Ta, Zr, Hf and other high field strength elements (HFSE) in magnetite in the Tashkurgan magnetite deposit in Xinjiang were depleted. Combined with the discrimination diagram of magnetite types, it was mainly divided into two genetic types: magmatic hydrothermal magnetite and hydrothermal metasomatic skarn magnetite related to marine volcanic activity. CONCLUSIONS The results show that the gypsum-salt formation changes the oxygen fugacity during the formation of magnetite deposits in the Tashkurgan area, and provides an important source of ore-forming material for mineralization during the formation of porphyrite-type iron ore in the Ningwu area.

Aiming to address the issue of complex preprocessing and post-processing required by existing crowd localization methods that employ pseudo bounding boxes or pre-designed localization maps, an end-to-end crowd localization network based on weakly supervised, LocalFormer, is proposed. In the feature extraction stage, a pure Transformer is used as the backbone network, and a global max-pooling operation is performed on the features of each stage to extract more comprehensive details of human heads. In the encoder-decoder stage, the positional information is embedded into the aggregated features as input to the encoder. Each decoder layer uses a set of trainable embeddings as queries, and takes visual features of the last layer of the encoder as keys and values. The decoded features are then used to predict confidence scores. Finally, a binary module is introduced with an adaptive optimized threshold learner to precisely binarize the confidence maps. Extensive experiments on three datasets in different environments show that the proposed method achieves the best positioning performance. [ABSTRACT FROM AUTHOR]