no method matching iterate(::RGB{Float64}) in seeded_region_growing

[jrtwolski]

Hello,
I have difficulty running seeded_region_growing for an RGB/jpeg image:
img2 = load("data/test/car10.jpg")
seeds = [(CartesianIndex(550,450),1), (CartesianIndex(750,250),2), (CartesianIndex(350,450),3)]
segments = seeded_region_growing(img2,seeds)

ends with:
ERROR: LoadError: MethodError: no method matching iterate(::RGB{Float64})
Closest candidates are:
iterate(::Union{LinRange, StepRangeLen}) at range.jl:664
iterate(::Union{LinRange, StepRangeLen}, ::Int64) at range.jl:664
iterate(::T) where T<:Union{Base.KeySet{var"#s77", var"#s76"} where {var"#s77", var"#s76"<:Dict}, Base.ValueIterator{var"#s75"} where var"#s75"<:Dict} at dict.jl:693
...
Stacktrace:
[1] _foldl_impl(op::Base.MappingRF{typeof(abs2), Base.BottomRF{typeof(Base.add_sum)}}, init::Base._InitialValue, itr::RGB{Float64})
@ Base .\reduce.jl:56
[2] foldl_impl(op::Base.MappingRF{typeof(abs2), Base.BottomRF{typeof(Base.add_sum)}}, nt::Base._InitialValue, itr::RGB{Float64})
@ Base .\reduce.jl:48
[3] mapfoldl_impl(f::typeof(abs2), op::typeof(Base.add_sum), nt::Base._InitialValue, itr::RGB{Float64})
@ Base .\reduce.jl:44
[4] mapfoldl(f::Function, op::Function, itr::RGB{Float64}; init::Base._InitialValue)
@ Base .\reduce.jl:160
[5] mapfoldl(f::Function, op::Function, itr::RGB{Float64})
@ Base .\reduce.jl:160
[6] mapreduce(f::Function, op::Function, itr::RGB{Float64}; kw::Base.Iterators.Pairs{Union{}, Union{}, Tuple{}, NamedTuple{(), Tuple{}}})
@ Base .\reduce.jl:287
[7] mapreduce(f::Function, op::Function, itr::RGB{Float64})
@ Base .\reduce.jl:287
[8] sum(f::Function, a::RGB{Float64}; kw::Base.Iterators.Pairs{Union{}, Union{}, Tuple{}, NamedTuple{(), Tuple{}}})
@ Base .\reduce.jl:501
[9] sum(f::Function, a::RGB{Float64})
@ Base .\reduce.jl:501
[10] default_diff_fn(c1::RGB{Float64}, c2::RGB{N0f8})
@ ImageSegmentation C:\Users\PC.julia\packages\ImageSegmentation\MX8ga\src\region_growing.jl:2
[11] seeded_region_growing(img::Matrix{RGB{N0f8}}, seeds::Vector{Tuple{CartesianIndex{2}, Int64}}, neighbourhood::ImageSegmentation.var"#20#24"{CartesianIndex{2}}, diff_fn::typeof(ImageSegmentation.default_diff_fn))
@ ImageSegmentation C:\Users\PC.julia\packages\ImageSegmentation\MX8ga\src\region_growing.jl:125
[12] seeded_region_growing(img::Matrix{RGB{N0f8}}, seeds::Vector{Tuple{CartesianIndex{2}, Int64}}, kernel_dim::Tuple{Int64, Int64}, diff_fn::Function)
@ ImageSegmentation C:\Users\PC.julia\packages\ImageSegmentation\MX8ga\src\region_growing.jl:64
[13] seeded_region_growing (repeats 2 times)
@ C:\Users\PC.julia\packages\ImageSegmentation\MX8ga\src\region_growing.jl:57 [inlined]

Any ideas what is wrong?

[johnnychen94]

Similar to JuliaImages/ImageSegmentation.jl#83. It's because Images isn't yet compatible with the latest ImageSegmentation version.

We're almost ready; if we go smoothly it should be released in two weeks I think.

[JaimeMAnayaRojas]

Hi, I am having trouble using the ImageSegmentation.jl:

# multiple objects
img = load("sample-images/kittens-555822_640.jpg");
imshow(img);

seeds = [
    (CartesianIndex(130,90), 1), 
    (CartesianIndex(130,200), 2), 
    (CartesianIndex(130,300), 2), 
    (CartesianIndex(130,420), 3), 
    (CartesianIndex(30,50), 4)
]

segments = seeded_region_growing(img, seeds)

This is the error message:

ERROR: MethodError: no method matching iterate(::RGB{Float64})
Closest candidates are:
iterate(::Union{LinRange, StepRangeLen}) at /opt/julia-1.7.2/share/julia/base/range.jl:826
iterate(::Union{LinRange, StepRangeLen}, ::Integer) at /opt/julia-1.7.2/share/julia/base/range.jl:826
iterate(::T) where T<:Union{Base.KeySet{<:Any, <:Dict}, Base.ValueIterator{<:Dict}} at /opt/julia-1.7.2/share/julia/base/dict.jl:695

...
Not sure what to do...

[spinnerwheel]

This problem is fixed in version 0.25.0

You could try removing Images:
( xxx) pkg> rm Images

And reinstalling it specifying the version you want:
( xxx) pkg> add Images@0.25.2

(In my case, there was a package, ImageFeatures, preventing me from installing the newer version, so I removed even that package)

I think there is a way better method to fix this issue, but hope that this would help someone 😃