Abstract: The present disclosure includes a method, apparatus and system for generating a neural network and a non-transitory computer readable storage medium storing instructions. The method comprises: recognizing at least an attribute of an object in a sample image according to a feature extracted from the sample image, using the neural network; determining a loss function value at least according to a margin value determined based on a semantic relationship between attributes, wherein the semantic relationship is obtained from a predefined table at least according to a real attribute and the recognized attribute of the object, wherein the predefined table is composed of the attributes and the semantic relationship between the attributes; updating a parameter in the neural network according to the determined loss function value. When using the neural network generated according the present disclosure, the accuracy of object attribute recognition can be improved.

Abstract: The present disclosure includes a method, apparatus and system for generating a neural network and a non-transitory computer readable storage medium storing instructions. The method comprises: recognizing at least an attribute of an object in a sample image according to a feature extracted from the sample image, using the neural network; determining a loss function value at least according to a margin value determined based on a semantic relationship between attributes, wherein the semantic relationship is obtained from a predefined table at least according to a real attribute and the recognized attribute of the object, wherein the predefined table is composed of the attributes and the semantic relationship between the attributes; updating a parameter in the neural network according to the determined loss function value. When using the neural network generated according the present disclosure, the accuracy of object attribute recognition can be improved.

Abstract: An attribute recognition apparatus including a unit for extracting a first feature from an image by using a feature extraction neural network; a unit for recognizing a first attribute of an object in the image based on the first feature by using a first recognition neural network; a unit for determining a second recognition neural network from a plurality of second recognition neural network candidates based on the first attribute; and a unit for recognizing at least one second attribute of the object based on the first feature by using a second recognition neural network.

Abstract: A multi-sensor based automatic brewing mass overturn machine comprising a moving module, a vinegar brewing mass information monitoring module, a system control module and a brewing mass overturn execution module is provided. The moving module comprises a moving motor, and the vinegar brewing mass information monitoring module comprises a vinegar brewing mass temperature and acidity information monitoring unit, a vinegar brewing mass smell information monitoring unit, and an overturned brewing mass appearance information monitoring unit. The system control module comprises a computer and a programmable controller, the computer being connected with a temperature sensor, an acidity sensor, a gas sensor array, a color camera and the programmable controller, respectively. The brewing mass overturn execution module comprises an overturn motor and brewing mass overturn motors.

Abstract: A multi-sensor based automatic brewing mass overturn machine comprising a moving module, a vinegar brewing mass information monitoring module, a system control module and a brewing mass overturn execution module is provided. The moving module comprises a moving motor, and the vinegar brewing mass information monitoring module comprises a vinegar brewing mass temperature and acidity information monitoring unit, a vinegar brewing mass smell information monitoring unit, and an overturned brewing mass appearance information monitoring unit. The system control module comprises a computer and a programmable controller, the computer being connected with a temperature sensor, an acidity sensor, a gas sensor array, a color camera and the programmable controller, respectively. The brewing mass overturn execution module comprises an overturn motor and brewing mass overturn motors.

Abstract: A hyperspectral imaging light source system includes a light box (1). The light box (1) is a sealed cuboid. In the chamber of the box body a line-scanning camera (2), a beam splitting system (3), an electric control translation platform carrier (5) and the object to be detected (4) on the electric control translation platform carrier, an electric control translation platform screw (6), a linear light source box (7) and a photosensitive diode (8) are respectively provided. A linear light source controller (9), a step motor (10) and a computer (11) are provided outside the box body. The photosensitive diode (8) senses linear light source intensity variation and inputs a feedback signal to the linear light source controller (9). A halogen lamp (13) is installed in the linear light source box (7), and is connected with the linear light source controller (9).

Abstract: A hyperspectral imaging light source system includes a light box (1). The light box (1) is a sealed cuboid. In the chamber of the box body a line-scanning camera (2), a beam splitting system (3), an electric control translation platform carrier (5) and the object to be detected (4) on the electric control translation platform carrier, an electric control translation platform screw (6), a linear light source box (7) and a photosensitive diode (8) are respectively provided. A linear light source controller (9), a step motor (10) and a computer (11) are provided outside the box body. The photosensitive diode (8) senses linear light source intensity variation and inputs a feedback signal to the linear light source controller (9). A halogen lamp (13) is installed in the linear light source box (7), and is connected with the linear light source controller (9).