Comparison of length parameters extracted from the HVI and AFIS to predict yarn quality

Predicting yarn quality from fiber properties is crucial to the cotton breeding and the textile industries. Fiber length parameters are critical in this endeavor. The current High Volume Instruments (HVI) provide two length measurements, Upper Half Mean Length (UHML) and Uniformity Index (U.I.), where U.I. is the ratio of Mean Length (ML) to the UHML expressed as a percentage. UHML corresponds to approximately the 1.8% span length and ML the 7.8% span length. Those two parameters inform us of the length of the longest fibers but do not measure fiber length variation related to shorter fibers in the sample. On the other hand, the Advanced Fiber Information System (AFIS) measures the length of individual fibers (typically 9,000 to 15,000 fibers). We demonstrated that using the complete fiber length distribution from the AFIS to build yarn prediction models is beneficial. Nevertheless, it is a slower instrument than the HVI making its use in cotton classification impossible. Our goal is to investigate if using the complete HVI fibrogram or some parameters extracted from it could solve this problem. Reconstructed length distribution was developed from the HVI fibrogram. Length parameters calculated from the reconstructed length distribution were compared with the current HVI length parameters and calculated length parameters from the AFIS length distribution by number. The results show that length parameters from the reconstructed length distribution are promising and are better at predicting yarn properties than the current HVI length parameters and as well as calculated length parameters from the AFIS.