Evidences about the impact of commodity width on brand performance remain fragmented. The traditional measurement of commodity category width is characterized by a quantity of manual work and repetition. A fusion model of convolutional neural network (CNN) and long short-term memory (LSTM) is proposed to solve the issue. In order to assure the universality and applicability of the findings, a vast consumer data set covering two retailers and two good categories is used for the measurement. The calculation results show that a composite model has the higher extraction accuracy than any single model on the average, and CNN or LSTM model alone will lead to the lower accuracy and higher error value. Convolutional neural network model possesses of powerful feature extraction, and the accuracy capacity which can be improved by CNN-LSTM fusion model. The mentioned fusion opens a new way for the measurement of commodity width.

The tracking of raw materials, production, and circulation links is carried out using QR codes in cigarette products. Each link’s current data collection and correlation analysis are still in the research and development stage. When used in conjunction with the GD packing machine’s production process feature, the QR code reader serves as an acquisition element for the process of data acquisition and the associated design for the tobacco packets and boxes, and the software is used to associate the received data and correct process exceptions. The system satisfies the design requirements, according to the results of the field verification.

This study employs the EIQ analysis approach to quantitatively examine the real order data of the cigarette distribution center in an effort to further increase the efficiency of cigarette sorting and distribution. To further increase the accuracy and efficiency of sorting labor, a sorting system based on QR code information and an accurate identification algorithm is proposed with the use of statistical data and the associated planning scheme. According to the system simulation findings, the sorting efficiency of the system’s sorting line is 18570 pieces per hour. However, by adding packing equipment, the efficiency of the system can be raised by 61.55 percent and can reach 30000 pieces per hour. Additionally, it confirms the viability of the system design. The two-dimensional code’ properties determine which global threshold method the binarization algorithm chooses. The two-dimensional code’ coding principle is investigated, and a two-dimensional code decoding algorithm is developed. The results of the experiment indicate that the decoding algorithm can fulfill these requirements.

This paper mainly studies the design method of the hardware and software of the cigarette strip QR code decoding based on STM32 microprocessor. This paper expounds the principle of hardware circuit design with STM32F407 microprocessor as the core, analyzes the steps of analyzing the QR code decoding algorithm based on the modular programming idea of C language, and analyzes the result of cigarette-bar QR code decoding. The results of field application show that the system can meet the design requirements of cigarette QR code decoding and improve the recognition rate of two-dimensional code, which has high application value.