Image retrieval system

A method for generating an image retrieval system configured to rank a plurality of images of cargo from a dataset of images, in response to a query corresponding to an image of cargo of interest generated using penetrating radiation, the method comprising:

• obtaining a plurality of annotated training images comprising cargo, each of the training images being associated with an annotation indicating a type of the cargo in the training image;
• and
• training the image retrieval system by applying a deep learning algorithm to the obtained annotated training images, the training comprising:
• applying, to the annotated training images, a feature extraction convolutional neural network, CNN, comprising a plurality of convolutional layers to generate a tensor of image features, and
• applying, to the generated tensor , an aggregated generalized mean, AgGeM, pooling layer associated with image spatial information, the applying comprising:
• applying a generalized mean pooling layer, to generate a plurality of embedding vectors (f(p))p∈P∈, such that:
• (
• f
• (
• p
• )
• )
• p
• =
• [
• f
• 1
• (
• p
• )
• ,
• ..
• ,
• f
• K
• (
• p
• )
• ]
• with:
• f
• k
• (
• p
• )
• =
• (
• 1
• ❘
• "\[LeftBracketingBar]"
• 𝒳
• k
• ❘
• "\[RightBracketingBar]"
• ⁢
• ∑
• x
• ∈
• 𝒳
• k
• ⁢
• x
• p
• )
• 1
• p
• with:
• P is a set of positive integers p representing pooling parameters of the generalized mean pooling layer,
• the tensor =(k)k∈{1, . . . , K} having H×W activations for a feature map k∈{1, . . . , K}, the feature map resulting from the application of the feature extraction CNN on an training image, with H and W being respectively a height and a width of each of the feature maps,
• K is a number of feature maps in a last convolutional layer of the feature extraction CNN,
• x is a feature from the generated tensor , and
• |k| is a cardinal of k of the tensor , and
• aggregating the generated plurality of embedding vectors (f(p))p∈P∈ by applying weights α of a scoring layer associated with an attention mechanism to the plurality of embedding vectors (f(p))p∈P∈, for each pooling parameter p belonging to P, the weights α being such that:
• α
• ⁡
• (
• f
• (
• p
• )
• ;
• θ
• )
• with the weights α and a parameter θ being learnable by the image retrieval system to minimize an associated loss function,
• wherein the aggregating is configured to generate a feature vector f such that:
• f
• =
• [
• f
• 1
• ,
• ..
• ,
• f
• K
• ]
• with
• ∀
• k
• ∈
• {
• 1
• ,
• ..
• ,
• K
• }
• ,
• f
• k
• =
• ∑
• p
• ∈
• 𝒫
• ⁢
• α
• ⁡
• (
• f
• (
• p
• )
• ;
• θ
• )
• ·
• f
• k
• (
• p
• )
• .